Biotech OutLook
Anna Huynh
7/16/2021
Dataset:
Dataset 01 covers the years 2015 to 2019 (though 2020 is included it is rare) for 464 Biotech companies that are publicly traded in the United States.
URL: https://www.kaggle.com/andersward/10k-statement-data-for-us-biotech-companies
Dataset 02 demonstrates 05 well-known biotech companies through movements of financial market in US, including:
- PFIZER: Pfizer Inc. develops, manufactures, and sells healthcare products worldwide. It offers medicines and vaccines in various therapeutic areas.
- ASTRAZENECA: Moderna, Inc., a clinical stage biotechnology company, develops therapeutics and vaccines based on messenger RNA for the treatment of infectious diseases, immuno-oncology, rare diseases, and cardiovascular diseases.
- BIONTECH: BioNTech SE, a biotechnology company, develops and commercializes immunotherapies for cancer and other infectious diseases.
- MODERNA: Moderna, Inc., a clinical stage biotechnology company, develops therapeutics and vaccines based on messenger RNA for the treatment of infectious diseases, immuno-oncology, rare diseases, and cardiovascular diseases.
- NOVAVAX: Novavax, Inc., together with its subsidiary, Novavax AB, a late-stage biotechnology company, focuses on the discovery, development, and commercialization of vaccines to prevent serious infectious diseases.
URL: https://www.kaggle.com/patrickgomes/pharmaceutical-companies-on-stock-exchange-in-2020
dt1 <- read.csv("C:/Users/Anhuynh/Desktop/Data Science Project/BioTech/biotech_timeSeries.csv")
dt2 <- read.csv("C:/Users/Anhuynh/Desktop/Data Science Project/BioTech/Pharmaceutical companies on the stock exchange in 2020.csv")Dataset 01: Explorartory Data Analysis
# Check missing data
mis_dt1 <- function(x) {sum(is.na(x))/length(x)*100}
apply(dt1, 2, mis_dt1)## year
## 0.00000
## symbol
## 0.00000
## marketCap
## 83.33333
## NetIncomeLoss
## 57.83046
## OperatingIncomeLoss
## 61.17098
## ResearchAndDevelopmentExpense
## 63.75718
## GeneralAndAdministrativeExpense
## 68.49856
## Revenues
## 86.99713
## CostOfGoodsSold
## 96.08477
## SalesRevenueGoodsNet
## 96.08477
## SalesRevenueNet
## 96.55172
## PaymentsToAcquirePropertyPlantAndEquipment
## 64.43966
## NetCashProvidedByUsedInOperatingActivities
## 71.73132
## CommonStockValue
## 82.00431
## CashAndCashEquivalentsAtCarryingValue
## 69.68391
## operatingMargin
## 55.35201library(mice)##
## Attaching package: 'mice'## The following object is masked from 'package:stats':
##
## filter## The following objects are masked from 'package:base':
##
## cbind, rbind# Impute data
imp1 <- mice(dt1, m=3, seed = 123, method = "cart")## Warning: Number of logged events: 4new_dt1 <- complete(imp1, 1)
imp2 <- mice(new_dt1, m=3, seed = 1234, method = "cart")## Warning: Number of logged events: 32dt1_imp <- complete(imp2, 1)
imp3 <- mice(dt1_imp, m=3, seed = 12345, method = "cart")## Warning: Number of logged events: 1dt1_impute <- complete(imp3, 1)# count on number of distinct symbols (#companies)
uni <- unique(dt1_impute$symbol)
length(uni)## [1] 464library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(ggplot2)
library(scales)
library(stringr)
# Sorting data for creating graph chart.
subMar = dt1_impute[order(dt1_impute$marketCap, decreasing = TRUE), ]
sort_subMar <- head(subMar, 5)
Mar <- sort_subMar[,c(2,3)]
Mar <- Mar %>%
mutate(marketCap_rounded = marketCap/10^3) %>%
group_by(symbol) %>%
mutate(Percent = (marketCap/sum(Mar$marketCap))*100)
subInc = dt1_impute[order(dt1_impute$NetIncomeLoss, decreasing = TRUE), ]
sort_subInc <- head(subInc, 5)
Inc <- sort_subInc[,c(2,4)]
Inc <- Inc %>%
mutate(NetIncomeLoss_rounded = NetIncomeLoss/10^9) %>%
group_by(symbol) %>%
mutate(Percent = (NetIncomeLoss/sum(Inc$NetIncomeLoss))*100)
subRev = dt1_impute[order(dt1_impute$Revenues, decreasing = TRUE), ]
sort_subRev <- head(subRev, 5)
Rev <- sort_subRev[,c(2,8)]
Rev <- Rev %>%
mutate(Revenues_rounded = Revenues/10^9) %>%
group_by(symbol) %>%
mutate(Percent = (Revenues/sum(Rev$Revenues))*100)
subStock = dt1_impute[order(dt1_impute$CommonStockValue, decreasing = TRUE), ]
sort_subStock <- head(subStock, 5)
Stock <- sort_subStock[,c(2,14)]
Stock <- Stock %>%
mutate(StockValue_rounded = CommonStockValue/10^6) %>%
group_by(symbol) %>%
mutate(Percent = (CommonStockValue/sum(Stock$CommonStockValue))*100)
subMargin = dt1_impute[order(dt1_impute$operatingMargin, decreasing = TRUE), ]
sort_subMargin <- head(subMargin, 5)
Margin <- sort_subMargin[,c(2,16)]
Margin <- Margin %>%
group_by(symbol) %>%
mutate(Percent = (operatingMargin/sum(Margin$operatingMargin))*100)
subRD = dt1_impute[order(dt1_impute$ResearchAndDevelopmentExpense, decreasing = TRUE), ]
sort_subRD <- head(subRD, 5)
RD <- sort_subRD[,c(2,6)]
RD <- RD %>%
mutate(RD_rounded = ResearchAndDevelopmentExpense/10^9) %>%
group_by(symbol) %>%
mutate(Percent = (ResearchAndDevelopmentExpense/sum(RD$ResearchAndDevelopmentExpense))*100)
# Data visualization
ggplot(Mar, aes(x="", y=Percent, fill=symbol))+
geom_bar(width = 0.5,stat ="identity")+
coord_polar(theta = "y",direction = 1)+
geom_text(aes(x=1.3,y = Percent, label = paste(round(Mar$Percent,0),'\n', "(", gsub(" ", "", paste("$",round(Mar$marketCap_rounded,1),"K")), ")") ), position = position_stack(vjust = 0.5)) +
scale_fill_brewer(palette="GnBu")+ggtitle("Market Cap Value")+theme_minimal() + labs(x = NULL)
ggplot(Inc, aes(x="", y=Percent, fill=symbol))+
geom_bar(width = 0.5,stat ="identity")+
coord_polar(theta = "y",direction = 1)+
geom_text(aes(x=1.3,y = Percent, label = paste(round(Inc$Percent,0),'\n', "(", gsub(" ", "", paste("$",round(Inc$NetIncomeLoss_rounded,1),"B")),")") ), position = position_stack(vjust = 0.5)) +
scale_fill_brewer(palette="GnBu")+ggtitle("Net Income Loss")+theme_minimal() +
labs(x = NULL)
ggplot(Rev, aes(x="", y=Percent, fill=symbol))+
geom_bar(width = 0.5,stat ="identity")+
coord_polar(theta = "y",direction = 1)+
geom_text(aes(x=1.3,y = Percent, label = paste(round(Rev$Percent,0),'\n', "(", gsub(" ", "", paste("$",round(Rev$Revenues_rounded,1),"B")),")") ), position = position_stack(vjust = 0.5)) +
scale_fill_brewer(palette="GnBu")+ggtitle("Revenues")+theme_minimal() +
labs(x = NULL)
ggplot(RD, aes(x="", y=Percent, fill=symbol))+
geom_bar(width = 0.5,stat ="identity")+
coord_polar(theta = "y",direction = 1)+
geom_text(aes(x=1.3,y = Percent, label = paste(round(RD$Percent,0),'\n', "(", gsub(" ", "", paste("$",round(RD$RD_rounded,1),"B")),")") ), position = position_stack(vjust = 0.5)) +
scale_fill_brewer(palette="GnBu")+ggtitle("R&D Expenses")+theme_minimal() +
labs(x = NULL)
ggplot(Stock, aes(x="", y=Percent, fill=symbol))+
geom_bar(width = 0.5,stat ="identity")+
coord_polar(theta = "y",direction = 1)+
geom_text(aes(x=1.3,y = Percent, label = paste(round(Stock$Percent,0),'\n', "(", gsub(" ", "", paste("$",round(Stock$StockValue_rounded,1), "M")),")") ), position = position_stack(vjust = 0.5)) +
scale_fill_brewer(palette="GnBu")+ggtitle("Common Stock Value")+ theme_minimal() + labs(x = NULL)
ggplot(Margin, aes(x="", y=Percent, fill=symbol))+
geom_bar(width = 0.5,stat ="identity")+
coord_polar(theta = "y",direction = 1)+
geom_text(aes(x=1.3,y = Percent, label = paste(round(Margin$Percent,0),'\n', "(", gsub(" ", "", paste("$",round(Margin$operatingMargin,1))),")") ), position = position_stack(vjust = 0.5)) +
scale_fill_brewer(palette="GnBu")+ggtitle("Operating Margin")+theme_minimal() + labs(x = NULL)
library(dplyr)
library(ggplot2)
library(scales)
library(stringr)
expenses <- dt1_impute[,c(1,6,7,9,12)] %>%
group_by(year) %>%
summarise('R&D' = sum(ResearchAndDevelopmentExpense), GeneralAndAdministrativeExpense = sum(GeneralAndAdministrativeExpense),
CostOfGoodsSold = sum(CostOfGoodsSold), PaymentsToAcquirePropertyPlantAndEquipment = sum(PaymentsToAcquirePropertyPlantAndEquipment))
# create new data for graph
df1 <- expenses[,c(1,2)]
colnames(df1) <- c("year", "value")
df1 <- df1 %>% mutate(category = "R&D")
df2 <- expenses[,c(1,3)]
colnames(df2) <- c("year", "value")
df2 <- df2 %>% mutate(category = "General&AdministrativeExpense")
df3 <- expenses[,c(1,4)]
colnames(df3) <- c("year", "value")
df3 <- df3 %>% mutate(category = "CostOfGoodsSold")
df4 <- expenses[,c(1,5)]
colnames(df4) <- c("year", "value")
df4 <- df4 %>% mutate(category = "PaymentsToAcquirePropertyPlant&Equipment")
df_cb <- rbind(df1, df2, df3, df4)
# data viz
ggplot(df_cb, aes(fill=category, y=value, x=year, label = paste(round(df_cb$value/10^9,1)))) +
geom_bar(position="stack", stat="identity") +
geom_text(position = position_stack(vjust = 0.5), size = 3, color = "black") + scale_fill_brewer(palette="GnBu") +
labs(title="Spending of Biotech Companies in the U.S (2015-2020)", x="Year",y="Expenses") +
scale_y_continuous(labels =
label_number(scale = 1e-9, prefix = "$", suffix = "B", big.mark = ',', accuracy = 1))## Warning: Use of `df_cb$value` is discouraged. Use `value` instead.
## Warning: Use of `df_cb$value` is discouraged. Use `value` instead.
# create data for visualization
sum1 = sum(df_cb$value[df_cb$year == '2015'])
sum2 = sum(df_cb$value[df_cb$year == '2016'])
sum3 = sum(df_cb$value[df_cb$year == '2017'])
sum4 = sum(df_cb$value[df_cb$year == '2018'])
sum5 = sum(df_cb$value[df_cb$year == '2019'])
sum6 = sum(df_cb$value[df_cb$year == '2020'])
viz2 <- expenses[, c(1,2)] %>%
mutate(total_exp = c(sum1, sum2, sum3, sum4, sum5, sum6))
df5 <- viz2[,c(1,2)]
colnames(df5) <- c("year", "value")
df5 <- df5 %>% mutate(category = "R&D")
df6 <- viz2[,c(1,3)]
colnames(df6) <- c("year", "value")
df6 <- df6 %>% mutate(category = "total_expense")
df_cb2 <- rbind(df5, df6)
# data viz2
ggplot(df_cb2, aes(fill=category, y=value, x=year, label = paste(round(df_cb2$value/10^9,1))) ) +
geom_bar(position="stack", stat="identity") +
geom_text(position = position_stack(), vjust = 0.5, size = 3, color = "black")+ scale_fill_brewer(palette="GnBu") +
labs(title="R&D Vs. Total Expenses of Biotech Companies (2015-2020)", x="Year",y="") +
scale_y_continuous(labels =
label_number(scale = 1e-9, prefix = "$", suffix = "B", big.mark = ',', accuracy = 1))## Warning: Use of `df_cb2$value` is discouraged. Use `value` instead.## Warning: Use of `df_cb2$value` is discouraged. Use `value` instead.
# data viz3
RE <- dt1_impute[,c(1,6,8)] %>%
group_by(year) %>%
summarise('R&D' = sum(ResearchAndDevelopmentExpense), Revenues = sum(Revenues))
# create new data for graph3
df7 <- RE[,c(1,2)]
colnames(df7) <- c("year", "value")
df7 <- df7 %>% mutate(category = "R&D")
df8 <- RE[,c(1,3)]
colnames(df8) <- c("year", "value")
df8 <- df8 %>% mutate(category = "Revenues")
df_cb3 <- rbind(df7, df8)
ggplot(df_cb3, aes(fill=category, y=value, x=year, label = paste(round(df_cb3$value/10^9,1))) ) +
geom_bar(position="stack", stat="identity") +
geom_text(position = position_stack(), vjust = 0.5, size = 3, color = "black")+ scale_fill_brewer(palette="GnBu") +
labs(title="R&D Vs. Revenues of Biotech Companies (2015-2020)", x="Year",y="") +
scale_y_continuous(labels =
label_number(scale = 1e-9, prefix = "$", suffix = "B", big.mark = ',', accuracy = 1))## Warning: Use of `df_cb3$value` is discouraged. Use `value` instead.## Warning: Use of `df_cb3$value` is discouraged. Use `value` instead.
# Compute % of deviation of R&D vs. Total Expenses
viz2_table <- viz2 %>%
mutate(RD_percent = (viz2$total_exp - viz2$`R&D`)/viz2$total_exp*100)
viz2_table## # A tibble: 6 x 4
## year `R&D` total_exp RD_percent
## <int> <dbl> <dbl> <dbl>
## 1 2015 58224239803 124138710065 53.1
## 2 2016 83532459692 141586913840 41.0
## 3 2017 78378233158 162151867375 51.7
## 4 2018 49187724053 100248797818 50.9
## 5 2019 68501523762 115979870386 40.9
## 6 2020 80600876946 156883496498 48.6# Compute % of deviation of R&D vs. Revenues
viz3_table <- RE %>%
mutate(RD_percent = (RE$Revenues - RE$`R&D`)/RE$Revenues*100)
viz3_table## # A tibble: 6 x 4
## year `R&D` Revenues RD_percent
## <int> <dbl> <dbl> <dbl>
## 1 2015 58224239803 203908403809 71.4
## 2 2016 83532459692 152409600862 45.2
## 3 2017 78378233158 244884325213 68.0
## 4 2018 49187724053 105909017040 53.6
## 5 2019 68501523762 154735402558 55.7
## 6 2020 80600876946 264238945192 69.5Dataset 02: Explorartory Data Analysis
# Check missing data
mis_dt2 <- function(x) {sum(is.na(x))/length(x)*100}
apply(dt2, 2, mis_dt2)## Companies PFIZER PFIZER.1 PFIZER.2 PFIZER.3
## 0 0 0 0 0
## PFIZER.4 PFIZER.5 PFIZER.6 ASTRAZENECA ASTRAZENECA.1
## 0 0 0 0 0
## ASTRAZENECA.2 ASTRAZENECA.3 ASTRAZENECA.4 ASTRAZENECA.5 ASTRAZENECA.6
## 0 0 0 0 0
## BIONTECH BIONTECH.1 BIONTECH.2 BIONTECH.3 BIONTECH.4
## 0 0 0 0 0
## BIONTECH.5 BIONTECH.6 MODERNA MODERNA.1 MODERNA.2
## 0 0 0 0 0
## MODERNA.3 MODERNA.4 MODERNA.5 MODERNA.6 NOVAVAX
## 0 0 0 0 0
## NOVAVAX.1 NOVAVAX.2 NOVAVAX.3 NOVAVAX.4 NOVAVAX.5
## 0 0 0 0 0
## NOVAVAX.6
## 01. PFIZER
library(dplyr)
library(plotly)##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(ggplot2)
library(wesanderson)
# PFIZER
pfiz <- dt2[3:255,c(1,6,7,8)]
pfiz$Companies <- as.Date(pfiz$Companies)
pfiz$PFIZER.4 <- as.numeric(as.character(pfiz$PFIZER.4))
pfiz$PFIZER.5 <- as.numeric(as.character(pfiz$PFIZER.5))
pfiz$PFIZER.6 <- as.numeric(as.character(pfiz$PFIZER.6))
colnames(pfiz) <- c("date", "volume_pfiz", "adj.close_pfiz", "variation_pfiz")
plot_ly(data = pfiz,
x = ~ date,
y = ~ volume_pfiz,
type = "scatter",
mode = "line",
name = "Trading Volume of Pfizer") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(pfiz$volume_pfiz),
yend = max(pfiz$volume_pfiz) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(pfiz$volume_pfiz) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(pfiz$volume_pfiz) * 1.05,
showarrow = FALSE) %>%
layout(title = "Trading Volume of Pfizer Inc.",
yaxis = list(title = "Volume"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))plot_ly(data = pfiz,
x = ~ date,
y = ~ adj.close_pfiz,
type = "scatter",
mode = "line",
name = "Ajusted Close Price of Pfizer") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(pfiz$adj.close_pfiz),
yend = max(pfiz$adj.close_pfiz) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(pfiz$adj.close_pfiz) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(pfiz$adj.close_pfiz) * 1.05,
showarrow = FALSE) %>%
layout(title = "Adjusted Close Price of Pfizer Inc.",
yaxis = list(title = "Adjusted Close Price (US Dollars)"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))pfiz %>%
ggplot(aes(x = date, y = variation_pfiz)) +
geom_line(color = wes_palette('Moonrise2', n=1, 'discrete')) +
geom_smooth(se = F, lty = 2, color = wes_palette('BottleRocket1', 1)) +
labs(x = "Date", y = "Variation",
title = "Pfizer Stock Price Variation",
subtitle = "Smoothed Line in Red") +
cowplot::theme_cowplot()## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
library(dplyr)
library(tsbox)
library(TSstudio)##
## Attaching package: 'TSstudio'## The following object is masked from 'package:tsbox':
##
## ts_plot## Quantify impact of Covid-19 on share prices
pre_covid_pfiz <- pfiz %>%
dplyr::filter(date < as.Date("2020-04-01")) %>%
dplyr::arrange(date)
post_covid_pfiz <- pfiz %>%
dplyr::filter(date >= as.Date("2020-04-01")) %>%
dplyr::arrange(date)
## The series before the outbreak of the pandemic
# Step 01: convert the series into a 'ts' object first.
ts.obj_pfiz <- ts(pre_covid_pfiz$adj.close_pfiz, start = c(2020, 1), frequency = 12)
ts_plot(ts.obj_pfiz,
title = "Stock Value of Pfizer",
Ytitle = "Price per share (US Dollars)",
slider = TRUE)ts_seasonal(ts.obj = ts.obj_pfiz, type = "all")# Step 02: review the series correlation with its past lags using the ACF and PACF to select time-series models for seasonal data functions.
ts_cor(ts.obj = ts.obj_pfiz, lag.max = 36)Observation: we can see strong correlation between the series and the first and seasonal lags.
## Step 03: select time-series models for seasonal data
methods <- list(
auto_arima = list(method = "auto.arima",
notes = "Auto ARIMA"),
hw1 = list(method = "HoltWinters",
method_arg = NULL,
notes = "HoltWinters Model"),
hw2 = list(method = "HoltWinters",
method_arg = list(seasonal = "multiplicative"),
notes = "HW with multip. seasonality"),
tslm = list(method = "tslm",
method_arg = list(formula = input ~ trend + season),
notes = "tslm with trend and seasonal"))
# Backtesting is the time series equivalent of the machine learning cross-validation training approach. The idea here is simple - test each model with backtesting, and select the one that performed best, on average, on the different testing partition.
# For the backtesting, we will split the series into 6 testing partitions, each 12 months spaced by 3 months from each other.
train_method = list(partitions = 6,
sample.out = 12,
space = 3)
# train the models:
# Note that the forecast horizon is set the the length of the post_covid series. In addition we will set the MAPA as the error metric to evaluate the performance of the different models on the testing partitions:
md_pfiz <- train_model(input = ts.obj_pfiz,
methods = methods,
train_method = train_method,
horizon = nrow(post_covid_pfiz),
error = "MAPE")## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zoo## # A tibble: 4 x 7
## model_id model notes avg_mape avg_rmse `avg_coverage_8~ `avg_coverage_9~
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
## 1 tslm tslm tslm wi~ 0.0656 2.19 0.514 0.708
## 2 hw2 HoltWi~ HW with~ 0.0657 2.17 0.458 0.736
## 3 hw1 HoltWi~ HoltWin~ 0.0663 2.20 0.486 0.736
## 4 auto_ari~ auto.a~ Auto AR~ 0.0748 2.50 0.472 0.681# Quantify the Covid-19 impact
# select the tslm with trend and seasonal model (tslm) to calculate the Covid19 effect when it demonstrated the lowest error through backtesting
post_covid_pfiz$yhat <- as.numeric(md_pfiz$forecast$tslm$forecast$mean)
post_covid_pfiz$upper95 <- as.numeric(md_pfiz$forecast$tslm$forecast$upper[,2])
post_covid_pfiz$lower95 <- as.numeric(md_pfiz$forecast$tslm$forecast$lower[,2])
post_covid_pfiz$price_change <-
post_covid_pfiz$adj.close_pfiz - post_covid_pfiz$yhat
post_covid_pfiz## date volume_pfiz adj.close_pfiz variation_pfiz yhat upper95
## 1 2020-04-01 26709625 29.23673 -0.161289215 27.73603115 30.822415
## 2 2020-04-02 27186876 30.26808 0.920303345 27.84106770 30.927451
## 3 2020-04-03 32837370 30.97713 0.958255768 28.20082607 31.287210
## 4 2020-04-06 40503217 31.83351 0.616699219 27.18829441 30.274678
## 5 2020-04-07 24226295 30.94950 -1.318786621 27.24531155 30.331695
## 6 2020-04-08 25523980 31.86113 0.578746796 26.90290623 29.989290
## 7 2020-04-09 30714930 32.58860 0.341556549 27.17977314 30.266157
## 8 2020-04-13 18270141 32.35839 -0.777988434 27.04605427 30.132438
## 9 2020-04-14 21845309 33.55548 0.483871460 27.56552067 30.651904
## 10 2020-04-15 21312196 33.12268 -0.085388184 26.84008389 29.926467
## 11 2020-04-16 20657451 33.03981 -0.227706909 26.27056220 29.345644
## 12 2020-04-17 24404843 33.98827 0.388992310 26.71312526 29.788207
## 13 2020-04-20 22530620 33.22398 -0.284629822 25.87754326 29.026465
## 14 2020-04-21 23975233 32.80039 -0.151802063 25.98257980 29.131501
## 15 2020-04-22 20186103 33.38052 -0.673625946 26.34233818 29.491260
## 16 2020-04-23 20185365 33.78569 0.388992310 25.32980652 28.478728
## 17 2020-04-24 17359380 34.42107 0.512336731 25.38682365 28.535745
## 18 2020-04-27 23394057 35.29587 0.702083588 25.04441833 28.193340
## 19 2020-04-28 35666728 34.90912 -1.081592560 25.32128525 28.470207
## 20 2020-04-29 26518535 35.10250 -0.322578430 25.18756638 28.336488
## 21 2020-04-30 29450973 35.32349 0.341556549 25.70703278 28.855954
## 22 2020-05-01 22679867 34.66049 -0.199241638 24.98159599 28.130517
## 23 2020-05-04 19893512 34.64208 0.018978119 24.41207431 27.558784
## 24 2020-05-05 24123952 35.46163 -0.284629822 24.85463737 28.001347
## 25 2020-05-06 24047432 35.46163 -0.436431885 24.01905537 27.246603
## 26 2020-05-07 33546080 34.37350 -1.375709534 24.12409191 27.351639
## 27 2020-05-08 19901312 34.61531 -0.028465271 24.48385029 27.711398
## 28 2020-05-11 25620421 35.43372 0.806449890 23.47131863 26.698866
## 29 2020-05-12 19335314 34.74551 -0.721065521 23.52833576 26.755883
## 30 2020-05-13 24596882 34.45720 -0.246681213 23.18593044 26.413478
## 31 2020-05-14 22362507 35.26632 1.062618256 23.46279736 26.690345
## 32 2020-05-15 25460002 35.11752 -0.189750671 23.32907848 26.556626
## 33 2020-05-18 22639077 35.40582 -0.256168365 23.84854488 27.076092
## 34 2020-05-19 19476234 35.04312 -0.161289215 23.12310810 26.350656
## 35 2020-05-20 17504200 34.99662 -0.303604126 22.55358642 25.787600
## 36 2020-05-21 17652181 34.65250 -0.294116974 22.99614948 26.230163
## 37 2020-05-22 15781226 34.87571 0.370018005 22.16056747 25.481686
## 38 2020-05-26 23260094 34.86641 -0.151802063 22.26560402 25.586723
## 39 2020-05-27 25904158 34.79201 0.000000000 22.62536240 25.946481
## 40 2020-05-28 22521767 35.50812 0.322578430 21.61283073 24.933950
## 41 2020-05-29 48638200 35.51743 -0.075901031 21.66984787 24.990967
## 42 2020-06-01 76884979 32.97847 -0.037948608 21.32744255 24.648561
## 43 2020-06-02 42690900 33.62949 0.597721100 21.60430946 24.925428
## 44 2020-06-03 38751996 33.62949 -0.123340607 21.47059059 24.791709
## 45 2020-06-04 27013388 33.48998 0.284629822 21.99005699 25.311176
## 46 2020-06-05 36705655 33.47139 -0.237190247 21.26462021 24.585739
## 47 2020-06-08 30194149 34.02940 0.749526978 20.69509853 24.030862
## 48 2020-06-09 21675088 33.67599 -0.275142670 21.13766158 24.473425
## 49 2020-06-10 29557217 33.40628 -0.417457581 20.30207958 23.730492
## 50 2020-06-11 63166642 30.96963 -2.182165146 20.40711613 23.835529
## 51 2020-06-12 37902472 31.38814 -0.313091278 20.76687450 24.195287
## 52 2020-06-15 36247482 31.02543 0.056926727 19.75434284 23.182755
## 53 2020-06-16 55616945 31.06263 -0.180265427 19.81135998 23.239772
## 54 2020-06-17 27509822 31.21144 -0.028463364 19.46895466 22.897367
## 55 2020-06-18 32278855 30.91383 -0.028463364 19.74582157 23.174234
## 56 2020-06-19 43986266 31.08124 -0.256166458 19.61210270 23.040515
## 57 2020-06-22 23332503 30.79293 -0.208728790 20.13156910 23.559981
## 58 2020-06-23 37690408 30.47672 -0.407968521 19.40613232 22.834545
## 59 2020-06-24 30817695 29.93731 -0.493360519 18.83661064 22.287292
## 60 2020-06-25 27418651 30.05821 0.199241638 19.27917369 22.729855
## 61 2020-06-26 42153992 29.79781 -0.199241638 18.44359169 21.991775
## 62 2020-06-29 23956261 30.35582 0.170778275 18.54862823 22.096812
## 63 2020-06-30 25451676 30.41162 0.161289215 18.90838661 22.456570
## 64 2020-07-01 73503009 31.37884 -0.759010315 17.89585495 21.444038
## 65 2020-07-02 45270881 32.09496 -0.104362488 17.95287209 21.501055
## 66 2020-07-06 27150935 32.09496 -0.417457581 17.61046677 21.158650
## 67 2020-07-07 25148018 31.64855 -0.018974304 17.88733368 21.435517
## 68 2020-07-08 24256545 31.38814 -0.246677399 17.75361481 21.301798
## 69 2020-07-09 20363702 31.11843 -0.256168365 18.27308121 21.821264
## 70 2020-07-10 24459019 31.46254 0.161291122 17.54764442 21.095828
## 71 2020-07-13 56569129 32.74596 0.455406189 16.97812274 20.555622
## 72 2020-07-14 32854129 32.76457 -0.132827759 17.42068580 20.998185
## 73 2020-07-15 26294454 33.22028 -0.256168365 16.58510380 20.264317
## 74 2020-07-16 18421812 33.10868 0.000000000 16.69014034 20.369354
## 75 2020-07-17 23347365 33.71318 0.370018005 17.04989872 20.729112
## 76 2020-07-20 36815693 33.94569 -0.939277649 16.03736706 19.716580
## 77 2020-07-21 23646912 34.12239 -0.237194061 16.09438419 19.773597
## 78 2020-07-22 90705554 35.86153 0.313091278 15.75197887 19.431192
## 79 2020-07-23 59639536 35.72203 -0.777992249 16.02884579 19.708059
## 80 2020-07-24 33870501 35.02451 -0.351043701 15.89512691 19.574340
## 81 2020-07-27 27640307 34.91291 0.170780182 16.41459332 20.093807
## 82 2020-07-28 56393954 36.28934 0.455410004 15.68915653 19.368370
## 83 2020-07-29 37033660 36.51255 -0.037952423 15.11963485 18.834633
## 84 2020-07-30 32273796 36.38107 0.436431885 15.56219791 19.277197
## 85 2020-07-31 28503006 36.13690 0.189750671 14.72661591 18.546960
## 86 2020-08-03 31036400 36.01482 -0.246677399 14.83165245 18.651996
## 87 2020-08-04 29448549 36.05238 0.075901031 15.19141083 19.011755
## 88 2020-08-05 24604787 36.10873 -0.018974304 14.17887917 17.999223
## 89 2020-08-06 20922111 35.93969 -0.474380493 14.23589630 18.056240
## 90 2020-08-07 23097251 36.10873 0.142314911 13.89349098 17.713835
## 91 2020-08-10 19964552 36.05238 -0.151802063 14.17035789 17.990702
## 92 2020-08-11 21326847 35.48891 -0.730548859 14.03663902 17.856983
## 93 2020-08-12 19965817 35.99603 0.474380493 14.55610542 18.376449
## 94 2020-08-13 15340021 35.84578 0.094875336 13.83066864 17.651013
## 95 2020-08-14 12901276 35.74248 0.047439575 13.26114696 17.123185
## 96 2020-08-17 17494187 36.01482 0.294116974 13.70371002 17.565748
## 97 2020-08-18 10846398 36.02421 -0.085388184 12.86812801 16.838627
## 98 2020-08-19 15690055 35.93030 -0.208728790 12.97316456 16.943663
## 99 2020-08-20 21957455 36.36229 0.512336731 13.33292294 17.303422
## 100 2020-08-21 25467170 36.51255 -0.028461456 12.32039127 16.290890
## 101 2020-08-24 25011947 36.47498 -0.170780182 12.37740841 16.347907
## 102 2020-08-25 26221201 36.07116 -0.094879150 12.03500309 16.005502
## 103 2020-08-26 24083689 35.73309 -0.132827759 12.31187000 16.282369
## 104 2020-08-27 22156556 35.55465 -0.094875336 12.17815113 16.148650
## 105 2020-08-28 32830519 35.60161 0.037952423 12.69761753 16.668116
## 106 2020-08-31 30032465 35.48891 -0.047439575 11.97218075 15.942679
## 107 2020-09-01 36161475 34.63433 -0.863376617 11.40265907 15.420230
## 108 2020-09-02 29068688 34.93484 0.407970428 11.84522212 15.862793
## 109 2020-09-03 35956788 34.18356 -0.920303345 11.00964012 15.138333
## 110 2020-09-04 26872257 34.14599 -0.123340607 11.11467667 15.243369
## 111 2020-09-08 26736818 33.74218 -0.569259644 11.47443504 15.603128
## 112 2020-09-09 27490533 33.97696 -0.104362488 10.46190338 14.590596
## 113 2020-09-10 21650741 33.47922 -0.521823883 10.51892052 14.647613
## 114 2020-09-11 24109301 33.87365 0.313095093 10.17651520 14.305208
## 115 2020-09-14 29565543 34.75641 0.616699219 10.45338211 14.582075
## 116 2020-09-15 22003726 34.70946 -0.218215942 10.31966324 14.448356
## 117 2020-09-16 22208412 34.54042 -0.474380493 10.83912964 14.967822
## 118 2020-09-17 17894390 34.57798 0.037948608 10.11369286 14.242385
## 119 2020-09-18 31694623 34.39955 -0.142314911 9.54417117 13.724820
## 120 2020-09-21 26608968 33.82669 -0.332069397 9.98673423 14.167383
## 121 2020-09-22 22727297 34.04269 0.417457581 9.15115223 13.445190
## 122 2020-09-23 23102521 33.80791 -0.322578430 9.25618877 13.550226
## 123 2020-09-24 20859925 33.58253 -0.180263519 9.61594715 13.909985
## 124 2020-09-25 14365598 33.85487 0.483871460 8.60341549 12.897453
## 125 2020-09-28 15044269 34.17416 0.218215942 8.66043262 12.954470
## 126 2020-09-29 13512491 33.96756 -0.246681213 8.31802731 12.612065
## 127 2020-09-30 22629275 34.46529 0.407966614 8.59489422 12.888932
## 128 2020-10-01 22520396 34.15538 -0.512332916 8.46117535 12.755213
## 129 2020-10-02 21461759 34.16477 0.303604126 8.98064175 13.274679
## 130 2020-10-05 19400029 34.51224 0.208728790 8.25520496 12.549242
## 131 2020-10-06 25233392 33.96756 -0.787475586 7.68568328 12.036107
## 132 2020-10-07 22160034 34.24929 0.199237823 8.12824634 12.478670
## 133 2020-10-08 17857395 34.64372 0.275142670 7.29266434 11.758403
## 134 2020-10-09 23667465 34.54981 -0.037948608 7.39770088 11.863440
## 135 2020-10-12 19413837 34.57798 0.113853455 7.75745926 12.223198
## 136 2020-10-13 21543233 34.65311 0.218215942 6.74492760 11.210667
## 137 2020-10-14 22124619 34.61555 -0.142314911 6.80194473 11.267684
## 138 2020-10-15 16089205 34.32442 -0.075904846 6.45953941 10.925278
## 139 2020-10-16 42993503 35.63917 0.834915161 6.73640633 11.202145
## 140 2020-10-19 31931352 35.49831 -0.588237762 6.60268745 11.068426
## 141 2020-10-20 22384852 35.20718 -0.360530853 7.12215385 11.587893
## 142 2020-10-21 18984859 34.82215 -0.256164551 6.39671707 10.862456
## 143 2020-10-22 19243616 35.15084 0.332065582 5.82719539 10.353337
## 144 2020-10-23 30557884 35.85517 0.237190247 6.26975845 10.795900
## 145 2020-10-26 33783124 35.61100 0.047439575 5.43417645 10.077269
## 146 2020-10-27 36851529 35.15084 0.370018005 5.53921299 10.182305
## 147 2020-10-28 33857537 33.29140 -1.280834198 5.89897137 10.542064
## 148 2020-10-29 27545341 33.13175 -0.189754486 4.88643970 9.529532
## 149 2020-10-30 25631067 33.31958 0.227703094 4.94345684 9.586549
## 150 2020-11-02 21430666 34.02391 0.284629822 4.60105152 9.244144
## 151 2020-11-03 21636617 33.98634 -0.227706909 4.87791843 9.521011
## 152 2020-11-04 40870536 35.05693 0.256168365 4.74419956 9.387292
## 153 2020-11-05 33926995 34.52562 -0.815940857 5.26366596 9.906758
## 154 2020-11-06 23870676 34.53510 0.009487152 4.53822918 9.181321
## 155 2020-11-09 230153864 37.19165 -2.523719788 3.96870750 8.675845
## 156 2020-11-10 80091668 36.69829 -1.679317474 4.41127055 9.118408
## 157 2020-11-11 58980997 36.52752 -0.360530853 3.57568855 8.401162
## 158 2020-11-12 46767877 35.62619 -0.692600250 3.68072510 8.506199
## 159 2020-11-13 40175634 36.64137 0.711574554 4.04048347 8.865957
## 160 2020-11-16 75530167 35.41746 -0.502845764 3.02795181 7.853426
## 161 2020-11-17 46984500 36.04000 -0.450000763 3.08496895 7.910443
## 162 2020-11-18 63442700 36.32000 -0.610000610 2.74256363 7.568037
## 163 2020-11-19 43261800 36.19000 -0.409999847 3.01943054 7.844904
## 164 2020-11-20 60633800 36.70000 0.000000000 2.88571167 7.711186
## 165 2020-11-23 43673100 36.52000 -0.509998322 3.40517807 8.230652
## 166 2020-11-24 42485300 36.60000 0.469997406 2.67974129 7.505215
## 167 2020-11-25 26955800 36.53000 0.000000000 2.11021961 7.003046
## 168 2020-11-27 25277000 37.23000 0.450000763 2.55278266 7.445609
## 169 2020-11-30 65425900 38.31000 0.360000610 1.71720066 6.729536
## 170 2020-12-01 72660800 39.41000 0.009998322 1.82223721 6.834572
## 171 2020-12-02 84347500 40.80000 0.329998016 2.18199558 7.194331
## 172 2020-12-03 68570100 40.09000 -0.889999390 1.16946392 6.181799
## 173 2020-12-04 35368100 40.34000 0.450000763 1.22648106 6.238816
## 174 2020-12-07 48253700 41.25000 0.590000153 0.88407574 5.896411
## 175 2020-12-08 87039300 42.56000 1.159999847 1.16094265 6.173278
## 176 2020-12-09 86118500 41.85000 -1.220001221 1.02722378 6.039559
## 177 2020-12-10 57749200 41.73000 -0.090000153 1.54669018 6.559025
## 178 2020-12-11 60737000 41.12000 -0.850002289 0.82125340 5.833588
## 179 2020-12-14 94809700 39.21000 -2.409999847 0.25173171 5.334424
## 180 2020-12-15 66003300 38.71000 -0.350002289 0.69429477 5.776987
## 181 2020-12-16 56515300 37.84000 -0.340000153 -0.14128723 5.061906
## 182 2020-12-17 52036400 38.03000 0.199996948 -0.03625069 5.166943
## 183 2020-12-18 60259200 37.68000 -0.310001373 0.32350769 5.526701
## 184 2020-12-21 40891800 37.38000 0.100002289 -0.68902397 4.514169
## 185 2020-12-22 33634400 36.74000 -0.469997406 -0.63200684 4.571186
## 186 2020-12-23 36182000 37.44000 0.430000305 -0.97441216 4.228781
## 187 2020-12-24 14790100 37.27000 -0.130001068 -0.69754524 4.505648
## 188 2020-12-28 26993700 36.82000 -0.540000916 -0.83126411 4.371929
## 189 2020-12-29 23152100 37.05000 0.149997711 -0.31179771 4.891396
## 190 2020-12-30 24889800 36.74000 -0.289997101 -1.03723450 4.165959
## 191 2020-12-31 30755600 36.81000 0.150001526 -1.60675618 3.669529
## lower95 price_change
## 1 24.64964770 1.500701
## 2 24.75468425 2.427008
## 3 25.11444262 2.776299
## 4 24.10191096 4.645213
## 5 24.15892810 3.704186
## 6 23.81652278 4.958227
## 7 24.09338969 5.408823
## 8 23.95967082 5.312333
## 9 24.47913722 5.989956
## 10 23.75370043 6.282601
## 11 23.19548009 6.769248
## 12 23.63804314 7.275148
## 13 22.72862181 7.346433
## 14 22.83365835 6.817809
## 15 23.19341673 7.038178
## 16 22.18088507 8.455884
## 17 22.23790220 9.034246
## 18 21.89549688 10.251453
## 19 22.17236380 9.587837
## 20 22.03864492 9.914931
## 21 22.55811132 9.616461
## 22 21.83267454 9.678896
## 23 21.26536496 10.230004
## 24 21.70792801 10.606991
## 25 20.79150796 11.442573
## 26 20.89654451 10.249409
## 27 21.25630288 10.131457
## 28 20.24377122 11.962405
## 29 20.30078836 11.217171
## 30 19.95838304 11.271272
## 31 20.23524995 11.803522
## 32 20.10153108 11.788437
## 33 20.62099748 11.557278
## 34 19.89556070 11.920009
## 35 19.31957314 12.443030
## 36 19.76213619 11.656354
## 37 18.83944859 12.715146
## 38 18.94448514 12.600805
## 39 19.30424352 12.166649
## 40 18.29171185 13.895291
## 41 18.34872899 13.847578
## 42 18.00632367 11.651027
## 43 18.28319058 12.025177
## 44 18.14947171 12.158895
## 45 18.66893811 11.499926
## 46 17.94350133 12.206766
## 47 17.35933527 13.334298
## 48 17.80189833 12.538326
## 49 16.87366721 13.104201
## 50 16.97870375 10.562515
## 51 17.33846213 10.621267
## 52 16.32593047 11.271090
## 53 16.38294760 11.251274
## 54 16.04054228 11.742483
## 55 16.31740920 11.168010
## 56 16.18369032 11.469133
## 57 16.70315672 10.661358
## 58 15.97771994 11.070588
## 59 15.38592904 11.100701
## 60 15.82849209 10.779039
## 61 14.89540840 11.354214
## 62 15.00044495 11.807191
## 63 15.36020332 11.503233
## 64 14.34767166 13.482986
## 65 14.40468880 14.142083
## 66 14.06228348 14.484489
## 67 14.33915039 13.761214
## 68 14.20543152 13.634527
## 69 14.72489792 12.845354
## 70 13.99946114 13.914899
## 71 13.40062320 15.767841
## 72 13.84318626 15.343886
## 73 12.90589053 16.635172
## 74 13.01092707 16.418537
## 75 13.37068545 16.663286
## 76 12.35815379 17.908323
## 77 12.41517092 18.028010
## 78 12.07276560 20.109551
## 79 12.34963252 19.693181
## 80 12.21591364 19.129386
## 81 12.73538004 18.498317
## 82 12.00994326 20.600184
## 83 11.40463622 21.392912
## 84 11.84719928 20.818871
## 85 10.90627189 21.410286
## 86 11.01130844 21.183168
## 87 11.37106681 20.860969
## 88 10.35853515 21.929847
## 89 10.41555229 21.703793
## 90 10.07314697 22.215236
## 91 10.35001388 21.882022
## 92 10.21629501 21.452275
## 93 10.73576141 21.439927
## 94 10.01032463 22.015107
## 95 9.39910875 22.481330
## 96 9.84167181 22.311110
## 97 8.89762944 23.156080
## 98 9.00266599 22.957133
## 99 9.36242436 23.029366
## 100 8.34989270 24.192155
## 101 8.40690984 24.097571
## 102 8.06450452 24.036160
## 103 8.34137143 23.421216
## 104 8.20765256 23.376502
## 105 8.72711896 22.903995
## 106 8.00168218 23.516734
## 107 7.38508818 23.231668
## 108 7.82765123 23.089619
## 109 6.88094761 23.173915
## 110 6.98598415 23.031316
## 111 7.34574253 22.267741
## 112 6.33321087 23.515052
## 113 6.39022801 22.960301
## 114 6.04782269 23.697134
## 115 6.32468960 24.303030
## 116 6.19097073 24.389794
## 117 6.71043713 23.701291
## 118 5.98500034 24.464287
## 119 5.36352230 24.855380
## 120 5.80608536 23.839960
## 121 4.85711482 24.891534
## 122 4.96215137 24.551722
## 123 5.32190975 23.966580
## 124 4.30937808 25.251454
## 125 4.36639522 25.513731
## 126 4.02398990 25.649533
## 127 4.30085681 25.870392
## 128 4.16713794 25.694209
## 129 4.68660434 25.184130
## 130 3.96116756 26.257036
## 131 3.33525955 26.281877
## 132 3.77782261 26.121044
## 133 2.82692531 27.351054
## 134 2.93196186 27.152108
## 135 3.29172023 26.820521
## 136 2.27918857 27.908183
## 137 2.33620571 27.813602
## 138 1.99380039 27.864882
## 139 2.27066730 28.902765
## 140 2.13694843 28.895619
## 141 2.65641483 28.085030
## 142 1.93097805 28.425434
## 143 1.30105347 29.323642
## 144 1.74361653 29.585409
## 145 0.79108423 30.176824
## 146 0.89612078 29.611624
## 147 1.25587915 27.392433
## 148 0.24334749 28.245312
## 149 0.30036463 28.376119
## 150 -0.04204069 29.422855
## 151 0.23482622 29.108421
## 152 0.10110735 30.312727
## 153 0.62057375 29.261950
## 154 -0.10486303 29.996874
## 155 -0.73843039 33.222943
## 156 -0.29586733 32.287021
## 157 -1.24978531 32.951827
## 158 -1.14474876 31.945461
## 159 -0.78499038 32.600882
## 160 -1.79752205 32.389506
## 161 -1.74050491 32.955032
## 162 -2.08291023 33.577436
## 163 -1.80604332 33.170568
## 164 -1.93976219 33.814289
## 165 -1.42029579 33.114822
## 166 -2.14573257 33.920257
## 167 -2.78260637 34.419779
## 168 -2.34004331 34.677217
## 169 -3.29513443 36.592801
## 170 -3.19009789 37.587763
## 171 -2.83033951 38.618004
## 172 -3.84287117 38.920536
## 173 -3.78585403 39.113519
## 174 -4.12825935 40.365924
## 175 -3.85139244 41.399059
## 176 -3.98511131 40.822775
## 177 -3.46564491 40.183309
## 178 -4.19108170 40.298746
## 179 -4.83096022 38.958267
## 180 -4.38839716 38.015704
## 181 -5.34448054 37.981287
## 182 -5.23944399 38.066249
## 183 -4.87968562 37.356493
## 184 -5.89221728 38.069025
## 185 -5.83520014 37.372009
## 186 -6.17760546 38.414411
## 187 -5.90073855 37.967546
## 188 -6.03445742 37.651264
## 189 -5.51499102 37.361797
## 190 -6.24042780 37.777236
## 191 -6.88304096 38.416758# the estimated total stock value change between April and August 2020 as result of the pandemic
sum(post_covid_pfiz$price_change)## [1] 4090.168# Data viz
plot_ly() %>%
add_ribbons(x = post_covid_pfiz$date,
ymin = post_covid_pfiz$adj.close_pfiz,
ymax = post_covid_pfiz$yhat,
line = list(color = 'rgba(255, 0, 0, 0.05)'),
fillcolor = 'rgba(255, 0, 0, 0.6)',
name = "Estimated Change") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(pfiz$adj.close_pfiz),
yend = max(pfiz$adj.close_pfiz) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2020-01-01"),
y = max(pfiz$adj.close_pfiz) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-11-01"),
y = max(pfiz$adj.close_pfiz) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = paste("Estimated increase in", "<br>",
"stock price: ~$4,090",
sep = ""),
x = as.Date("2020-07-01"),
y = 2 * 10^1,
arrowhead = 1,
ax = -150,
ay = -10,
showarrow = TRUE) %>%
add_lines(x = pfiz$date,
y = pfiz$adj.close_pfiz,
line = list(color = "#1f77b4"),
name = "Actual") %>%
layout(title = "Covid19 Impact on Stock Value of Pfizer Inc.",
yaxis = list(title = "Stock Price (US Dollars)"),
xaxis = list(title = "Time Series Model - tslm with trend and seasonal",
range = c(as.Date("2019-10-01"), as.Date("2021-01-01"))),
legend = list(x = 0, y = 0.95))2. ASTRAZENECA
library(dplyr)
library(plotly)
library(ggplot2)
library(wesanderson)
# ASTRAZENECA
astra <- dt2[3:253,c(1,13,14,15)]
astra$Companies <- as.Date(astra$Companies)
astra$ASTRAZENECA.4 <- as.numeric(as.character(astra$ASTRAZENECA.4))
astra$ASTRAZENECA.5 <- as.numeric(as.character(astra$ASTRAZENECA.5))
astra$ASTRAZENECA.6 <- as.numeric(as.character(astra$ASTRAZENECA.6))
colnames(astra) <- c("date", "volume_astra", "adj.close_astra", "variation_astra")
plot_ly(data = astra,
x = ~ date,
y = ~ volume_astra,
type = "scatter",
mode = "line",
name = "Trading Volume of Astrazeneca") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(astra$volume_astra),
yend = max(astra$volume_astra) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(astra$volume_astra) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(astra$volume_astra) * 1.05,
showarrow = FALSE) %>%
layout(title = "Trading Volume of Astrazeneca Inc.",
yaxis = list(title = "Volume"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))plot_ly(data = astra,
x = ~ date,
y = ~ adj.close_astra,
type = "scatter",
mode = "line",
name = "Ajusted Close Price of Astrazeneca") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(astra$adj.close_astra),
yend = max(astra$adj.close_astra) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(astra$adj.close_astra) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(astra$adj.close_astra) * 1.05,
showarrow = FALSE) %>%
layout(title = "Adjusted Close Price of Astrazeneca Inc.",
yaxis = list(title = "Adjusted Close Price"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))astra %>%
ggplot(aes(x = date, y = variation_astra)) +
geom_line(color = wes_palette('Moonrise2', n=1, 'discrete')) +
geom_smooth(se = F, lty = 2, color = wes_palette('BottleRocket1', 1)) +
labs(x = "Date", y = "Variation",
title = "Astrazeneca Stock Price Variation",
subtitle = "Smoothed Line in Red") +
cowplot::theme_cowplot()## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
library(dplyr)
library(tsbox)
library(TSstudio)
## Quantify impact of Covid-19 on share prices
pre_covid_astra <- astra %>%
dplyr::filter(date < as.Date("2020-04-01")) %>%
dplyr::arrange(date)
post_covid_astra <- astra %>%
dplyr::filter(date >= as.Date("2020-04-01")) %>%
dplyr::arrange(date)
## The series before the outbreak of the pandemic
# Step 01: convert the series into a 'ts' object first.
ts.obj_astra <- ts(pre_covid_astra$adj.close_astra, start = c(2020, 1), frequency = 12)
ts_plot(ts.obj_astra,
title = "Stock Value of ASTRAZENECA",
Ytitle = "Price per share (US Dollars)",
slider = TRUE)ts_seasonal(ts.obj = ts.obj_astra, type = "all")# Step 02: review the series correlation with its past lags using the ACF and PACF to select time-series models for seasonal data functions.
ts_cor(ts.obj = ts.obj_astra, lag.max = 36)## Step 03: select time-series models for seasonal data
methods <- list(
auto_arima = list(method = "auto.arima",
notes = "Auto ARIMA"),
hw1 = list(method = "HoltWinters",
method_arg = NULL,
notes = "HoltWinters Model"),
hw2 = list(method = "HoltWinters",
method_arg = list(seasonal = "multiplicative"),
notes = "HW with multip. seasonality"),
tslm = list(method = "tslm",
method_arg = list(formula = input ~ trend + season),
notes = "tslm with trend and seasonal"))
# Backtesting is the time series equivalent of the machine learning cross-validation training approach. The idea here is simple - test each model with backtesting, and select the one that performed best, on average, on the different testing partition.
# For the backtesting, we will split the series into 6 testing partitions, each 12 months spaced by 3 months from each other.
train_method = list(partitions = 6,
sample.out = 12,
space = 3)
# train the models:
# Note that the forecast horizon is set the the length of the post_covid series. In addition we will set the MAPA as the error metric to evaluate the performance of the different models on the testing partitions:
md_astra <- train_model(input = ts.obj_astra,
methods = methods,
train_method = train_method,
horizon = nrow(post_covid_astra),
error = "MAPE")## # A tibble: 4 x 7
## model_id model notes avg_mape avg_rmse `avg_coverage_8~ `avg_coverage_9~
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
## 1 hw1 HoltWi~ HoltWin~ 0.0821 3.80 0.389 0.528
## 2 hw2 HoltWi~ HW with~ 0.0822 3.81 0.389 0.486
## 3 tslm tslm tslm wi~ 0.0864 3.99 0.194 0.375
## 4 auto_ari~ auto.a~ Auto AR~ 0.0955 4.43 0.361 0.431# Quantify the Covid-19 impact
# select the HoltWinters Model (hw1) to calculate the Covid-19 effect when it demonstrated the lowest error through backtesting
post_covid_astra$yhat <- as.numeric(md_astra$forecast$hw1$forecast$mean)
post_covid_astra$upper95 <- as.numeric(md_astra$forecast$hw1$forecast$upper[,2])
post_covid_astra$lower95 <- as.numeric(md_astra$forecast$hw1$forecast$lower[,2])
post_covid_astra$price_change <-
post_covid_astra$adj.close_astra - post_covid_astra$yhat
post_covid_astra## date volume_astra adj.close_astra variation_astra yhat upper95
## 1 2020-04-01 6097700 43.43278 0.02999878 43.12983 46.07910
## 2 2020-04-02 5918800 43.41294 0.32999802 43.40092 47.05644
## 3 2020-04-03 3372700 43.61135 0.18999863 42.93923 47.18511
## 4 2020-04-06 5735200 43.95858 0.29000092 43.31835 48.08197
## 5 2020-04-07 5600600 42.75817 -0.38000107 43.25865 48.48903
## 6 2020-04-08 3860200 43.50223 0.43999863 42.89990 48.55865
## 7 2020-04-09 5628200 44.30581 0.75000000 43.19489 49.25179
## 8 2020-04-13 6555700 46.98439 1.95000076 42.95786 49.38832
## 9 2020-04-14 5109000 48.29392 1.04000092 43.87225 50.65572
## 10 2020-04-15 2901800 47.64907 0.36000061 43.53043 50.64943
## 11 2020-04-16 3218600 48.74035 0.58000183 43.38274 50.82214
## 12 2020-04-17 3349400 49.63321 -0.17000198 43.51511 51.26168
## 13 2020-04-20 3416400 50.03996 0.61999893 42.38282 50.55395
## 14 2020-04-21 3545400 49.67290 -0.02000046 42.65392 51.10566
## 15 2020-04-22 2803600 49.96060 -0.27999878 42.19222 50.91556
## 16 2020-04-23 2549400 49.99036 0.09999847 42.57134 51.55806
## 17 2020-04-24 2087600 50.77409 0.47999954 42.51165 51.75426
## 18 2020-04-27 2105000 51.07172 -0.13000107 42.15289 51.64449
## 19 2020-04-28 5767500 50.56576 -1.34999847 42.44788 52.18211
## 20 2020-04-29 3786000 50.31775 -1.29999924 42.21086 52.18180
## 21 2020-04-30 5863600 51.86537 -1.69000244 43.12525 53.32743
## 22 2020-05-01 3206400 51.44870 -0.16999817 42.78343 53.21171
## 23 2020-05-04 2852900 51.93481 0.12999725 42.63573 53.28532
## 24 2020-05-05 4579300 53.04593 0.84000015 42.76811 53.63449
## 25 2020-05-06 6604600 53.61141 -0.02000046 41.63582 52.80884
## 26 2020-05-07 2626600 52.32172 -0.75000000 41.90691 53.28676
## 27 2020-05-08 1660100 52.86736 0.19000244 41.44522 53.02821
## 28 2020-05-11 3095300 53.52213 0.92000198 41.82433 53.60696
## 29 2020-05-12 2221500 53.68086 -0.31999969 41.76464 53.74358
## 30 2020-05-13 3431200 54.87134 0.49000168 41.40589 53.57797
## 31 2020-05-14 4998300 53.07570 -0.49000168 41.70088 54.06309
## 32 2020-05-15 3716800 52.98641 0.82999802 41.46385 54.01331
## 33 2020-05-18 4926300 53.02609 -0.18999863 42.37824 55.11219
## 34 2020-05-19 3271200 53.14514 0.18000031 42.03642 54.95223
## 35 2020-05-20 4038000 53.35347 -0.55000305 41.88873 54.98387
## 36 2020-05-21 20428200 54.84158 -1.67000198 42.02110 55.29316
## 37 2020-05-22 6031200 54.78205 0.51000214 40.88881 54.41306
## 38 2020-05-26 5626600 52.68879 -1.40999985 41.15991 54.85553
## 39 2020-05-27 7010200 52.05386 0.52000046 40.69821 54.56308
## 40 2020-05-28 4445800 53.07570 -0.11999893 41.07733 55.10941
## 41 2020-05-29 7615800 54.16697 -0.68000031 41.01764 55.21496
## 42 2020-06-01 5759800 54.72253 0.84000015 40.65888 55.01954
## 43 2020-06-02 4279400 54.28602 0.22999954 40.95387 55.47604
## 44 2020-06-03 5464200 54.56380 0.45000076 40.71685 55.39874
## 45 2020-06-04 4483700 53.44276 -0.72000122 41.63124 56.47113
## 46 2020-06-05 6092400 53.42292 -0.49000168 41.28942 56.28566
## 47 2020-06-08 9762600 52.16299 -0.22999954 41.14172 56.29269
## 48 2020-06-09 4782300 52.31180 -0.06999969 41.27409 56.57823
## 49 2020-06-10 4582600 53.21458 -0.22000122 40.14181 55.66516
## 50 2020-06-11 4787000 51.16100 -2.20000076 40.41290 56.08578
## 51 2020-06-12 5630300 51.06179 -0.70999908 39.95121 55.77220
## 52 2020-06-15 5711800 52.22252 0.20000076 40.33032 56.29805
## 53 2020-06-16 5535500 53.15506 -0.59999847 40.27063 56.38376
## 54 2020-06-17 4919100 53.47252 -0.21999741 39.91188 56.16911
## 55 2020-06-18 3796100 52.24236 0.02000046 40.20687 56.60694
## 56 2020-06-19 6807200 52.59950 0.52000046 39.96984 56.51151
## 57 2020-06-22 3375400 52.99633 0.04999924 40.88423 57.56630
## 58 2020-06-23 10772000 53.81975 1.08000183 40.54241 57.36372
## 59 2020-06-24 6418600 51.71656 -0.78999710 40.39472 57.35411
## 60 2020-06-25 4948300 52.58958 0.51999664 40.52709 57.62345
## 61 2020-06-26 6234700 52.28204 -0.38000107 39.39480 56.68768
## 62 2020-06-29 4526200 52.49038 -0.45999908 39.66590 57.09312
## 63 2020-06-30 4621900 52.47053 0.61000061 39.20420 56.76475
## 64 2020-07-01 5462000 52.89712 0.48999786 39.58332 57.27618
## 65 2020-07-02 4591300 53.37331 0.16999817 39.52363 57.34783
## 66 2020-07-06 5180900 53.67094 0.25000000 39.16487 57.11945
## 67 2020-07-07 3747500 53.04593 -0.02999878 39.45986 57.54387
## 68 2020-07-08 6247800 53.76022 0.47999954 39.22283 57.43536
## 69 2020-07-09 4203000 53.54197 -0.52999878 40.13722 58.47736
## 70 2020-07-10 5414500 53.33363 -0.55000305 39.79541 58.26228
## 71 2020-07-13 5639800 52.81776 -0.68999863 39.64771 58.24045
## 72 2020-07-14 7199600 53.55189 0.47999954 39.78008 58.49785
## 73 2020-07-15 19902500 57.54000 1.68000031 38.64780 57.54522
## 74 2020-07-16 12358700 56.97453 0.47000122 38.91889 57.93933
## 75 2020-07-17 24319500 60.61542 3.44999695 38.45720 57.59987
## 76 2020-07-20 56346600 58.21461 -4.00999832 38.83631 58.10044
## 77 2020-07-21 15419100 57.60945 -1.58000183 38.77662 58.16144
## 78 2020-07-22 9562100 56.52810 -0.40000153 38.41787 57.92263
## 79 2020-07-23 8658900 54.73245 -1.58000183 38.71286 58.33683
## 80 2020-07-24 7748000 55.35745 0.48999786 38.47583 58.21830
## 81 2020-07-27 5660200 56.13127 0.42000198 39.39022 59.25047
## 82 2020-07-28 4979100 55.87333 -0.56000137 39.04840 59.02575
## 83 2020-07-29 5038900 56.19079 -0.40999985 38.90070 58.99446
## 84 2020-07-30 8215100 57.43088 0.77000046 39.03308 59.24257
## 85 2020-07-31 6844500 55.33761 -2.04000092 37.90079 58.27680
## 86 2020-08-03 4799700 56.36937 0.13999939 38.17188 58.66203
## 87 2020-08-04 4121300 55.63523 0.00000000 37.71019 58.31385
## 88 2020-08-05 3267200 55.50626 -0.47999954 38.08931 58.80586
## 89 2020-08-06 3532100 55.76420 0.11999893 38.02961 58.85845
## 90 2020-08-07 3849200 54.97055 -0.36999893 37.67086 58.61137
## 91 2020-08-10 2937300 54.94078 -0.31999969 37.96585 59.01744
## 92 2020-08-11 3948900 54.74237 -0.54000092 37.72882 58.89092
## 93 2020-08-12 5466200 56.29000 0.44000244 38.64321 59.91523
## 94 2020-08-13 3490600 55.80000 -0.54999924 38.30139 59.68278
## 95 2020-08-14 2893300 55.19000 -0.28000259 38.15370 59.64389
## 96 2020-08-17 2997600 56.46000 0.52000046 38.28607 59.88452
## 97 2020-08-18 4174800 56.33000 -0.50000000 37.15378 58.90811
## 98 2020-08-19 3322100 56.23000 -0.90999985 37.42488 59.28615
## 99 2020-08-20 2527500 56.49000 0.21000290 36.96318 58.93089
## 100 2020-08-21 3419200 55.71000 0.21999741 37.34230 59.41592
## 101 2020-08-24 7379000 56.76000 -0.74000168 37.28261 59.46164
## 102 2020-08-25 3399700 57.01000 -0.15999985 36.92385 59.20779
## 103 2020-08-26 2802400 57.00000 0.06999969 37.21885 59.60720
## 104 2020-08-27 4195100 56.11000 -0.88999939 36.98182 59.47411
## 105 2020-08-28 3251900 55.69000 -0.50000000 37.89621 60.49196
## 106 2020-08-31 3262600 56.00000 0.20000076 37.55439 60.25312
## 107 2020-09-01 3965900 55.38000 -0.46999741 37.40669 60.20795
## 108 2020-09-02 3614200 55.90000 0.43000031 37.53907 60.44238
## 109 2020-09-03 5122000 54.16000 -1.18999863 36.40678 59.45715
## 110 2020-09-04 4610000 53.58000 -0.20999908 36.67787 59.82921
## 111 2020-09-08 6735600 54.71000 0.56999969 36.21618 59.46804
## 112 2020-09-09 11416300 53.64000 -0.19000244 36.59529 59.94725
## 113 2020-09-10 4407100 53.07000 -1.09000015 36.53560 59.98722
## 114 2020-09-11 4627700 53.73000 -0.26000214 36.17685 59.72771
## 115 2020-09-14 4960000 54.02000 -0.20999908 36.47184 60.12153
## 116 2020-09-15 4300900 55.28000 -0.25000000 36.23481 59.98291
## 117 2020-09-16 3948500 55.13000 -0.59000015 37.14920 60.99531
## 118 2020-09-17 3991900 55.94000 0.19999695 36.80738 60.75110
## 119 2020-09-18 2931700 56.45000 -0.30999756 36.65969 60.70062
## 120 2020-09-21 6792700 55.91000 0.81999969 36.79206 60.92981
## 121 2020-09-22 3878600 55.44000 0.14999771 35.65977 59.93710
## 122 2020-09-23 2922900 55.66000 -0.56000137 35.93087 60.30408
## 123 2020-09-24 4123700 54.17000 -0.77000046 35.46917 59.93789
## 124 2020-09-25 4954900 55.51000 1.30999756 35.84829 60.41214
## 125 2020-09-28 2741400 54.75000 -0.31999969 35.78860 60.44721
## 126 2020-09-29 4177500 55.11000 0.40000153 35.42984 60.18286
## 127 2020-09-30 3001400 54.80000 -0.22000122 35.72483 60.57190
## 128 2020-10-01 4728200 54.30000 -0.90000153 35.48781 60.42856
## 129 2020-10-02 5699100 54.22000 -0.19999695 36.40220 61.43629
## 130 2020-10-05 6704300 54.49000 0.03000259 36.06038 61.18747
## 131 2020-10-06 3695000 53.71000 -0.44000244 35.91268 61.13242
## 132 2020-10-07 3518700 53.53000 -0.28000259 36.04505 61.35711
## 133 2020-10-08 2351300 54.09000 0.40999985 34.91277 60.35796
## 134 2020-10-09 2092000 54.72000 0.33000183 35.18386 60.72055
## 135 2020-10-12 2783500 55.27000 0.08000183 34.72217 60.35003
## 136 2020-10-13 2949800 54.48000 -0.29000092 35.10128 60.81999
## 137 2020-10-14 4657400 53.19000 -0.81000137 35.04159 60.85082
## 138 2020-10-15 3038700 52.55000 0.09999847 34.68284 60.58228
## 139 2020-10-16 2665500 53.03000 0.02999878 34.97783 60.96717
## 140 2020-10-19 3143700 52.44000 -0.79000092 34.74080 60.81973
## 141 2020-10-20 4919100 52.19000 -0.49000168 35.65519 61.82340
## 142 2020-10-21 8219800 51.57000 -0.56999969 35.31337 61.57056
## 143 2020-10-22 3751000 51.96000 -0.04999924 35.16568 61.51154
## 144 2020-10-23 5576900 52.00000 -0.16999817 35.29805 61.73229
## 145 2020-10-26 6022000 53.07000 0.20000076 34.16576 60.72752
## 146 2020-10-27 3702300 52.48000 -0.83000183 34.43686 61.08628
## 147 2020-10-28 5263200 50.98000 -0.90999985 33.97516 60.71196
## 148 2020-10-29 3534000 51.07000 0.02999878 34.35428 61.17816
## 149 2020-10-30 4761600 50.16000 -0.75999832 34.29459 61.20528
## 150 2020-11-02 5368800 50.62000 0.14999771 33.93583 60.93306
## 151 2020-11-03 4156800 51.70000 0.24000168 34.23082 61.31430
## 152 2020-11-04 7385800 55.04000 1.09000015 33.99379 61.16325
## 153 2020-11-05 6660900 55.16000 -0.90000153 34.90818 62.16335
## 154 2020-11-06 3686300 55.52000 -0.16999817 34.56637 61.90697
## 155 2020-11-09 4381300 54.05000 -0.72000122 34.41867 61.84445
## 156 2020-11-10 5835800 56.57000 -0.06999969 34.55104 62.06173
## 157 2020-11-11 4069700 58.02000 0.47999954 33.41876 61.05200
## 158 2020-11-12 2902900 56.84000 0.02999878 33.68985 61.40737
## 159 2020-11-13 2900500 57.38000 0.51000214 33.22816 61.02969
## 160 2020-11-16 5190500 56.29000 -0.20000076 33.60727 61.49257
## 161 2020-11-17 6451500 54.96000 -0.15000153 33.54758 61.51640
## 162 2020-11-18 4732300 53.99000 -1.34000015 33.18883 61.24091
## 163 2020-11-19 6768200 54.03000 -0.22000122 33.48382 61.61892
## 164 2020-11-20 4306000 55.30000 0.41999817 33.24679 61.46467
## 165 2020-11-23 13670600 54.70000 0.18000031 34.16118 62.46159
## 166 2020-11-24 8341400 53.57000 -0.43000031 33.81936 62.20206
## 167 2020-11-25 11660900 52.60000 0.25999832 33.67167 62.13642
## 168 2020-11-27 14783200 52.61000 1.02999878 33.80404 62.35062
## 169 2020-11-30 14725200 52.94000 -0.36000061 32.67175 61.33645
## 170 2020-12-01 13308200 52.98000 -0.08000183 32.94284 61.68880
## 171 2020-12-02 4953700 53.39000 0.50000000 32.48115 61.30812
## 172 2020-12-03 7917100 52.78000 -0.09000015 32.86027 61.76803
## 173 2020-12-04 4685400 53.74000 0.19000244 32.80058 61.78891
## 174 2020-12-07 6582900 54.26000 -0.12000275 32.44182 61.51050
## 175 2020-12-08 6707600 54.72000 1.05000305 32.73681 61.88562
## 176 2020-12-09 7952400 54.04000 -0.91999817 32.49978 61.72850
## 177 2020-12-10 7566800 53.89000 0.02999878 33.41417 62.72257
## 178 2020-12-11 3870000 54.27000 0.63999939 33.07236 62.46023
## 179 2020-12-14 54898600 50.03000 -0.63999939 32.92466 62.39179
## 180 2020-12-15 22345300 50.83000 0.22000122 33.05703 62.60320
## 181 2020-12-16 19333500 50.93000 -0.86999893 31.92475 61.58506
## 182 2020-12-17 13957200 50.79000 -0.46999741 32.19584 61.93468
## 183 2020-12-18 25875200 50.21000 -0.70999908 31.73414 61.55131
## 184 2020-12-21 13007200 49.68000 0.20000076 32.11326 62.00855
## 185 2020-12-22 14221900 48.73000 -0.54000092 32.05357 62.02677
## 186 2020-12-23 11311000 48.77000 -0.34000015 31.69481 61.74573
## 187 2020-12-24 4931700 48.52000 -0.38000107 31.98981 62.11823
## 188 2020-12-28 15481900 49.38000 -0.84999847 31.75278 61.95852
## 189 2020-12-29 17308500 49.90000 -0.75999832 32.66717 62.95002
## lower95 price_change
## 1 40.180561 0.30295138
## 2 39.745402 0.01201741
## 3 38.693351 0.67212541
## 4 38.554719 0.64023496
## 5 38.028282 -0.50047933
## 6 37.241150 0.60232875
## 7 37.137986 1.11091477
## 8 36.527408 4.02652785
## 9 37.088787 4.42167045
## 10 36.411442 4.11864039
## 11 35.943337 5.35761418
## 12 35.768541 6.11810179
## 13 34.211702 7.65713856
## 14 34.202179 7.01897940
## 15 33.468890 7.76837420
## 16 33.584616 7.41902084
## 17 33.269038 8.26244503
## 18 32.661292 8.91882290
## 19 32.713659 8.11787676
## 20 32.239909 8.10688843
## 21 32.923068 8.74012155
## 22 32.355145 8.66527523
## 23 31.986143 9.29908163
## 24 31.901717 10.27782714
## 25 30.462802 11.97559347
## 26 30.527064 10.41481163
## 27 29.862230 11.42214482
## 28 30.041709 11.69779145
## 29 29.785707 11.91621565
## 30 29.233807 13.46545393
## 31 29.338670 11.37481586
## 32 28.914395 11.52255708
## 33 29.644290 10.64785182
## 34 29.120612 11.10871488
## 35 28.793583 11.46474647
## 36 28.749044 12.82047562
## 37 27.364562 13.89324195
## 38 27.464286 11.52888269
## 39 26.833341 11.35565099
## 40 27.045246 11.99836687
## 41 26.820315 13.14933196
## 42 26.298220 14.06364822
## 43 26.431709 13.33214864
## 44 26.034955 13.84695147
## 45 26.791337 11.81152538
## 46 26.293177 12.13349914
## 47 25.990752 11.02127282
## 48 25.969960 11.03770754
## 49 24.618456 13.07277703
## 50 24.740024 10.74810206
## 51 24.130217 11.11058738
## 52 24.362593 11.89219257
## 53 24.157500 12.88443192
## 54 23.654642 13.56064622
## 55 23.806797 12.03548743
## 56 23.428166 12.62966267
## 57 24.202154 12.11209699
## 58 23.721107 13.27733628
## 59 23.435325 11.32184443
## 60 23.430727 12.06249180
## 61 22.101928 12.88723795
## 62 22.238671 12.82448010
## 63 21.643655 13.26633019
## 64 21.890453 13.31380452
## 65 21.699427 13.84968783
## 66 21.210297 14.50606570
## 67 21.375853 13.58607048
## 68 21.010310 14.53738531
## 69 21.797085 13.40474547
## 70 21.328534 13.53822359
## 71 21.054970 13.17005016
## 72 21.062320 13.77180440
## 73 19.750374 18.89220867
## 74 19.898448 18.05563566
## 75 19.314525 22.15822184
## 76 19.572187 19.37830359
## 77 19.391803 18.83283085
## 78 18.913102 18.11022999
## 79 19.088880 16.01959192
## 80 18.733359 16.88162376
## 81 19.529963 16.74105317
## 82 19.071053 16.82492878
## 83 18.806948 17.29008741
## 84 18.823582 18.39780020
## 85 17.524785 17.43682172
## 86 17.681734 18.19748153
## 87 17.106528 17.92504482
## 88 17.372753 17.41695808
## 89 17.200784 17.73458730
## 90 16.730352 17.29968709
## 91 16.914259 16.97493342
## 92 16.566730 17.01354760
## 93 17.371193 17.64678835
## 94 16.920014 17.49860448
## 95 16.663513 17.03629958
## 96 16.687628 18.17392723
## 97 15.399460 19.17621703
## 98 15.563604 18.80512101
## 99 14.995482 19.52681729
## 100 15.268682 18.36769907
## 101 15.103581 19.47738913
## 102 14.639915 20.08614432
## 103 14.830487 19.78115460
## 104 14.489525 19.12818323
## 105 15.300458 17.79379169
## 106 14.855655 18.44561087
## 107 14.605440 17.97330765
## 108 14.635751 18.36093530
## 109 13.356406 17.75322068
## 110 13.526537 16.90212893
## 111 12.964318 18.49382033
## 112 13.243342 17.04470500
## 113 13.083987 16.53439613
## 114 12.625989 17.55315117
## 115 12.822154 17.54816068
## 116 12.486711 19.04518703
## 117 13.303092 17.98079976
## 118 12.863666 19.13261513
## 119 12.618759 19.79031297
## 120 12.654312 19.11793925
## 121 11.382443 19.78022509
## 122 11.557656 19.72913257
## 123 11.000457 18.70082504
## 124 11.284439 19.66170956
## 125 11.129981 18.96140207
## 126 10.676822 19.68015787
## 127 10.877768 19.07516509
## 128 10.547049 18.81219312
## 129 11.368099 17.81780554
## 130 10.933288 18.42962381
## 131 10.692942 17.79731693
## 132 10.733004 17.48494381
## 133 9.467575 19.17723224
## 134 9.647172 19.53613958
## 135 9.094309 20.54783296
## 136 9.382578 19.37871641
## 137 9.232360 18.14840632
## 138 8.783395 17.86716212
## 139 8.988489 18.05217026
## 140 8.661874 17.69919814
## 141 9.486983 16.53480858
## 142 9.056188 16.25662746
## 143 8.819816 16.79432255
## 144 8.863810 16.70195066
## 145 7.604006 18.90423741
## 146 7.787437 18.04314353
## 147 7.238367 17.00483768
## 148 7.530392 16.71572219
## 149 7.383892 15.86541317
## 150 6.938606 16.68416745
## 151 7.147344 17.46917788
## 152 6.824337 21.04620606
## 153 7.653018 20.25181543
## 154 7.225760 20.95363385
## 155 6.992891 19.63132834
## 156 7.040354 22.01895599
## 157 5.785516 24.60124381
## 158 5.972336 23.15014977
## 159 5.426623 24.15184483
## 160 5.721973 22.68272904
## 161 5.578767 21.41241804
## 162 5.136745 20.80117582
## 163 5.348717 20.54618152
## 164 5.028913 22.05321001
## 165 5.860769 20.53882197
## 166 5.436658 19.75063872
## 167 5.206907 18.92833320
## 168 5.257460 18.80596253
## 169 4.007051 20.26824760
## 170 4.196894 20.03715479
## 171 3.654178 20.90884878
## 172 3.952499 19.91973254
## 173 3.812239 20.93942626
## 174 3.373138 21.81817810
## 175 3.588005 21.98318959
## 176 3.271072 21.54021731
## 177 4.105774 20.47582623
## 178 3.684486 21.19764511
## 179 3.457533 17.10533914
## 180 3.510862 17.77296938
## 181 2.264434 19.00525491
## 182 2.456997 18.59416179
## 183 1.916979 18.47585411
## 184 2.217976 17.56673969
## 185 2.080370 16.67642976
## 186 1.643902 17.07518586
## 187 1.861381 16.53019446
## 188 1.547039 17.62722310
## 189 2.384313 17.23283399# the estimated total stock value change between April and August 2020 as result of the pandemic
sum(post_covid_astra$price_change)## [1] 2924.645# Data viz
plot_ly() %>%
add_ribbons(x = post_covid_astra$date,
ymin = post_covid_astra$adj.close_astra,
ymax = post_covid_astra$yhat,
line = list(color = 'rgba(255, 0, 0, 0.05)'),
fillcolor = 'rgba(255, 0, 0, 0.6)',
name = "Estimated Change") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(astra$adj.close_astra),
yend = max(astra$adj.close_astra) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2020-01-01"),
y = max(astra$adj.close_astra) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2020-11-01"),
y = max(astra$adj.close_astra) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = paste("Estimated increase in", "<br>",
"stock price: ~$3,000",
sep = ""),
x = as.Date("2020-09-01"),
y = 4 * 10^1,
arrowhead = 1,
ax = -120,
ay = 40,
showarrow = TRUE) %>%
add_lines(x = astra$date,
y = astra$adj.close_astra,
line = list(color = "#1f77b4"),
name = "Actual") %>%
layout(title = "Covid19 Impact on Stock Value of ASTRAZENECA",
yaxis = list(title = "Stock Price (US Dollars)"),
xaxis = list(title = "Time Series Model - HoltWinters Model",
range = c(as.Date("2019-10-01"), as.Date("2021-01-01"))),
legend = list(x = 0, y = 0.95))3. BIONTECH
library(dplyr)
library(plotly)
library(ggplot2)
library(wesanderson)
# BIONTECH
bio <- dt2[3:251,c(1,20,21,22)]
bio$Companies <- as.Date(bio$Companies)
bio$BIONTECH.4 <- as.numeric(as.character(bio$BIONTECH.4))
bio$BIONTECH.5 <- as.numeric(as.character(bio$BIONTECH.5))
bio$BIONTECH.6 <- as.numeric(as.character(bio$BIONTECH.6))
colnames(bio) <- c("date", "volume_bio", "adj.close_bio", "variation_bio")
plot_ly(data = bio,
x = ~ date,
y = ~ volume_bio,
type = "scatter",
mode = "line",
name = "Trading Volume of BIONTECH") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(bio$volume_bio),
yend = max(bio$volume_bio) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(bio$volume_bio) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(bio$volume_bio) * 1.05,
showarrow = FALSE) %>%
layout(title = "Trading Volume of BIONTECH Inc.",
yaxis = list(title = "Volume"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))plot_ly(data = bio,
x = ~ date,
y = ~ adj.close_bio,
type = "scatter",
mode = "line",
name = "Ajusted Close Price of BIONTECH") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(bio$adj.close_bio),
yend = max(bio$adj.close_bio) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(bio$adj.close_bio) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(bio$adj.close_bio) * 1.05,
showarrow = FALSE) %>%
layout(title = "Adjusted Close Price of BIONTECH Inc.",
yaxis = list(title = "Adjusted Close Price"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))bio %>%
ggplot(aes(x = date, y = variation_bio)) +
geom_line(color = wes_palette('Moonrise2', n=1, 'discrete')) +
geom_smooth(se = F, lty = 2, color = wes_palette('BottleRocket1', 1)) +
labs(x = "Date", y = "Variation",
title = "BIONTECH Stock Price Variation",
subtitle = "Smoothed Line in Red") +
cowplot::theme_cowplot()## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
library(dplyr)
library(tsbox)
library(TSstudio)
## Quantify impact of Covid-19 on share prices
pre_covid_bio <- bio %>%
dplyr::filter(date < as.Date("2020-04-01")) %>%
dplyr::arrange(date)
post_covid_bio <- bio %>%
dplyr::filter(date >= as.Date("2020-04-01")) %>%
dplyr::arrange(date)
## The series before the outbreak of the pandemic
# Step 01: convert the series into a 'ts' object first.
ts.obj_bio <- ts(pre_covid_bio$adj.close_bio, start = c(2020, 3), frequency = 12)
ts_plot(ts.obj_bio,
title = "Stock Value of BIONTECH",
Ytitle = "Price per share (US Dollars)",
slider = TRUE)ts_seasonal(ts.obj = ts.obj_bio, type = "all")# Step 02: review the series correlation with its past lags using the ACF and PACF to select time-series models for seasonal data functions.
ts_cor(ts.obj = ts.obj_bio, lag.max = 36)## Step 03: select time-series models for seasonal data
methods <- list(
auto_arima = list(method = "auto.arima",
notes = "Auto ARIMA"),
hw1 = list(method = "HoltWinters",
method_arg = NULL,
notes = "HoltWinters Model"),
hw2 = list(method = "HoltWinters",
method_arg = list(seasonal = "multiplicative"),
notes = "HW with multip. seasonality"),
tslm = list(method = "tslm",
method_arg = list(formula = input ~ trend + season),
notes = "tslm with trend and seasonal"))
# Backtesting is the time series equivalent of the machine learning cross-validation training approach. The idea here is simple - test each model with backtesting, and select the one that performed best, on average, on the different testing partition.
# For the backtesting, we will split the series into 6 testing partitions, each 12 months spaced by 3 months from each other.
train_method = list(partitions = 6,
sample.out = 12,
space = 3)
# train the models:
# Note that the forecast horizon is set the the length of the post_covid series. In addition we will set the MAPA as the error metric to evaluate the performance of the different models on the testing partitions:
md_bio <- train_model(input = ts.obj_bio,
methods = methods,
train_method = train_method,
horizon = nrow(post_covid_bio),
error = "MAPE")## # A tibble: 4 x 7
## model_id model notes avg_mape avg_rmse `avg_coverage_8~ `avg_coverage_9~
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
## 1 auto_ari~ auto.a~ Auto AR~ 0.220 16.4 0.556 0.653
## 2 hw2 HoltWi~ HW with~ 0.237 16.9 0.444 0.556
## 3 hw1 HoltWi~ HoltWin~ 0.239 16.9 0.458 0.583
## 4 tslm tslm tslm wi~ 0.294 19.5 0.194 0.403# Quantify the Covid-19 impact
# select the Auto ARIMA (auto_arima) to calculate the Covid-19 effect when it demonstrated the lowest error through back-testing
post_covid_bio$yhat <- as.numeric(md_bio$forecast$auto_arima$forecast$mean)
post_covid_bio$upper95 <- as.numeric(md_bio$forecast$auto_arima$forecast$upper[,2])
post_covid_bio$lower95 <- as.numeric(md_bio$forecast$auto_arima$forecast$lower[,2])
post_covid_bio$price_change <-
post_covid_bio$adj.close_bio - post_covid_bio$yhat
post_covid_bio## date volume_bio adj.close_bio variation_bio yhat upper95
## 1 2020-04-01 259100 52.92 -3.430000305 58.4 72.32269
## 2 2020-04-02 245700 56.91 1.409999847 58.4 78.08965
## 3 2020-04-03 227300 52.75 -5.229999542 58.4 82.51480
## 4 2020-04-06 331000 52.73 -2.770000458 58.4 86.24537
## 5 2020-04-07 1461700 45.60 -4.400001526 58.4 89.53208
## 6 2020-04-08 819800 44.95 -3.149997711 58.4 92.50348
## 7 2020-04-09 1483000 45.41 -6.599998474 58.4 95.23597
## 8 2020-04-13 880700 41.71 -4.310001373 58.4 97.77931
## 9 2020-04-14 863900 41.63 -2.189998627 58.4 100.16806
## 10 2020-04-15 980600 38.58 -2.319999695 58.4 102.42740
## 11 2020-04-16 849400 38.79 0.030002594 58.4 104.57633
## 12 2020-04-17 1008300 40.81 0.510002136 58.4 106.62960
## 13 2020-04-20 1278800 43.37 1.879997253 58.4 108.59896
## 14 2020-04-21 1507200 42.26 -3.010002136 58.4 110.49392
## 15 2020-04-22 10852500 53.50 -0.939998627 58.4 112.32233
## 16 2020-04-23 4271200 50.00 -8.000000000 58.4 114.09075
## 17 2020-04-24 1822700 47.75 -3.490001678 58.4 115.80471
## 18 2020-04-27 1183300 49.40 0.510002136 58.4 117.46896
## 19 2020-04-28 1511200 45.83 -2.769996643 58.4 119.08759
## 20 2020-04-29 1766400 49.25 1.250000000 58.4 120.66415
## 21 2020-04-30 1034100 47.48 -1.729999542 58.4 122.20177
## 22 2020-05-01 874600 45.11 -0.889999390 58.4 123.70319
## 23 2020-05-04 784900 45.78 0.579998016 58.4 125.17086
## 24 2020-05-05 2604800 50.00 1.590000153 58.4 126.60696
## 25 2020-05-06 889400 47.78 -2.360000610 58.4 128.01343
## 26 2020-05-07 815300 49.32 0.060001373 58.4 129.39205
## 27 2020-05-08 696100 49.24 0.080001831 58.4 130.74440
## 28 2020-05-11 1887400 49.98 0.369998932 58.4 132.07193
## 29 2020-05-12 1910200 49.33 -1.069999695 58.4 133.37596
## 30 2020-05-13 1468800 48.49 -2.009998322 58.4 134.65770
## 31 2020-05-14 1099100 48.66 0.099998474 58.4 135.91824
## 32 2020-05-15 793800 49.49 0.980003357 58.4 137.15861
## 33 2020-05-18 5013900 60.17 5.429996490 58.4 138.37975
## 34 2020-05-19 3612100 50.59 -6.409999847 58.4 139.58252
## 35 2020-05-20 1979300 53.04 -0.169998169 58.4 140.76773
## 36 2020-05-21 1094300 51.90 -2.199996948 58.4 141.93612
## 37 2020-05-22 941600 51.03 -1.120002747 58.4 143.08840
## 38 2020-05-26 1511100 50.88 -1.520000458 58.4 144.22521
## 39 2020-05-27 1810200 47.59 -1.939998627 58.4 145.34715
## 40 2020-05-28 1107900 46.87 -1.470001221 58.4 146.45480
## 41 2020-05-29 760600 49.53 1.340000153 58.4 147.54869
## 42 2020-06-01 1275800 52.30 2.299999237 58.4 148.62932
## 43 2020-06-02 688500 52.38 0.400001526 58.4 149.69716
## 44 2020-06-03 1379900 51.16 -0.689998627 58.4 150.75266
## 45 2020-06-04 623300 50.00 -0.450000763 58.4 151.79622
## 46 2020-06-05 768700 49.18 -0.619998932 58.4 152.82826
## 47 2020-06-08 1142400 48.78 -0.440002441 58.4 153.84913
## 48 2020-06-09 970100 47.79 -0.009998322 58.4 154.85920
## 49 2020-06-10 913100 47.41 -0.389999390 58.4 155.85881
## 50 2020-06-11 1324800 48.60 -0.400001526 58.4 156.84826
## 51 2020-06-12 858800 48.58 -0.789997101 58.4 157.82787
## 52 2020-06-15 1087500 50.85 2.129997253 58.4 158.79792
## 53 2020-06-16 901100 49.90 -1.099998474 58.4 159.75869
## 54 2020-06-17 723000 50.85 1.099998474 58.4 160.71044
## 55 2020-06-18 691500 50.94 1.079998016 58.4 161.65341
## 56 2020-06-19 735800 51.36 0.459999084 58.4 162.58785
## 57 2020-06-22 2678200 57.45 5.500000000 58.4 163.51398
## 58 2020-06-23 2142200 52.75 -2.869998932 58.4 164.43202
## 59 2020-06-24 1344500 54.73 1.729999542 58.4 165.34219
## 60 2020-06-25 2406400 59.71 5.200000763 58.4 166.24467
## 61 2020-06-26 2830700 62.13 5.500000000 58.4 167.13966
## 62 2020-06-29 4011400 65.80 0.120002747 58.4 168.02735
## 63 2020-06-30 1979100 66.74 0.579994202 58.4 168.90790
## 64 2020-07-01 9879300 64.14 -13.559997559 58.4 169.78149
## 65 2020-07-02 3926500 63.27 -4.975002289 58.4 170.64829
## 66 2020-07-06 1649400 65.51 -0.079994202 58.4 171.50844
## 67 2020-07-07 2264400 69.34 5.339996338 58.4 172.36210
## 68 2020-07-08 1449600 67.56 0.349998474 58.4 173.20942
## 69 2020-07-09 1186700 65.61 -1.000000000 58.4 174.05052
## 70 2020-07-10 2874300 70.36 3.360000610 58.4 174.88555
## 71 2020-07-13 8529600 77.78 -0.690002441 58.4 175.71464
## 72 2020-07-14 4748700 83.88 1.309997559 58.4 176.53791
## 73 2020-07-15 3908700 78.00 -8.080001831 58.4 177.35549
## 74 2020-07-16 2847900 75.87 -4.629997253 58.4 178.16748
## 75 2020-07-17 5053500 85.25 8.709999084 58.4 178.97400
## 76 2020-07-20 7790800 88.20 -3.547004700 58.4 179.77517
## 77 2020-07-21 6025500 91.60 -0.400001526 58.4 180.57108
## 78 2020-07-22 13817800 104.17 4.979995728 58.4 181.36184
## 79 2020-07-23 15899700 88.51 -9.003997803 58.4 182.14755
## 80 2020-07-24 7109700 84.17 -2.130004883 58.4 182.92830
## 81 2020-07-27 5129200 86.53 -0.910003662 58.4 183.70418
## 82 2020-07-28 5108600 85.38 -3.620002747 58.4 184.47529
## 83 2020-07-29 3846300 82.50 -3.040000916 58.4 185.24171
## 84 2020-07-30 3302800 84.16 5.110000610 58.4 186.00353
## 85 2020-07-31 2640500 82.11 -3.889999390 58.4 186.76083
## 86 2020-08-03 2431700 85.47 2.480003357 58.4 187.51369
## 87 2020-08-04 2410300 83.17 -1.270004272 58.4 188.26218
## 88 2020-08-05 4084400 81.30 -1.519996643 58.4 189.00638
## 89 2020-08-06 3175200 77.08 -2.209999084 58.4 189.74636
## 90 2020-08-07 3038200 77.00 -0.089996338 58.4 190.48220
## 91 2020-08-10 2830600 73.95 -3.050003052 58.4 191.21397
## 92 2020-08-11 4032200 68.45 -0.550003052 58.4 191.94172
## 93 2020-08-12 3252000 69.11 -1.779998779 58.4 192.66553
## 94 2020-08-13 2390900 71.00 2.750000000 58.4 193.38546
## 95 2020-08-14 1862400 68.64 -2.290000916 58.4 194.10156
## 96 2020-08-17 1948000 70.26 1.170005798 58.4 194.81391
## 97 2020-08-18 1905100 67.88 -1.810005188 58.4 195.52256
## 98 2020-08-19 2326600 65.60 -1.819999695 58.4 196.22757
## 99 2020-08-20 2932600 66.27 0.192993164 58.4 196.92898
## 100 2020-08-21 6290400 73.02 1.250000000 58.4 197.62687
## 101 2020-08-24 3393900 70.85 -6.139999390 58.4 198.32127
## 102 2020-08-25 1576200 70.49 0.419998169 58.4 199.01224
## 103 2020-08-26 1711400 67.90 -2.099998474 58.4 199.69984
## 104 2020-08-27 1980500 65.59 -2.610000610 58.4 200.38410
## 105 2020-08-28 1760000 64.00 -1.040000916 58.4 201.06509
## 106 2020-08-31 2217200 61.25 -1.569999695 58.4 201.74283
## 107 2020-09-01 3569600 57.81 -0.899997711 58.4 202.41739
## 108 2020-09-02 3892800 61.99 5.980003357 58.4 203.08880
## 109 2020-09-03 2586000 59.28 -4.170001984 58.4 203.75712
## 110 2020-09-04 1985500 58.57 -0.709999084 58.4 204.42237
## 111 2020-09-08 2887800 59.83 -0.329998016 58.4 205.08461
## 112 2020-09-09 3406900 62.32 -0.680000305 58.4 205.74387
## 113 2020-09-10 1751800 61.20 -0.799999237 58.4 206.40019
## 114 2020-09-11 2515200 65.13 3.677997589 58.4 207.05362
## 115 2020-09-14 3373600 67.48 -0.019996643 58.4 207.70418
## 116 2020-09-15 3168300 66.92 -3.080001831 58.4 208.35192
## 117 2020-09-16 2231500 66.04 0.529998779 58.4 208.99688
## 118 2020-09-17 2860000 66.78 -0.040000916 58.4 209.63909
## 119 2020-09-18 2608900 67.00 -1.010002136 58.4 210.27858
## 120 2020-09-21 2118700 66.12 0.560005188 58.4 210.91539
## 121 2020-09-22 1572300 67.05 1.670005798 58.4 211.54955
## 122 2020-09-23 1678000 63.83 -3.419998169 58.4 212.18110
## 123 2020-09-24 2863400 62.83 0.210002899 58.4 212.81007
## 124 2020-09-25 1670500 66.69 5.420001984 58.4 213.43648
## 125 2020-09-28 1750000 66.50 -0.489997864 58.4 214.06037
## 126 2020-09-29 847600 66.86 0.260002136 58.4 214.68177
## 127 2020-09-30 1746800 69.23 1.830001831 58.4 215.30071
## 128 2020-10-01 2371700 72.71 2.709999084 58.4 215.91722
## 129 2020-10-02 2327800 73.70 1.849998474 58.4 216.53132
## 130 2020-10-05 3750600 80.70 5.000000000 58.4 217.14305
## 131 2020-10-06 5702100 85.59 -1.250000000 58.4 217.75243
## 132 2020-10-07 2354700 86.23 2.910003662 58.4 218.35949
## 133 2020-10-08 1826000 87.95 2.119995117 58.4 218.96426
## 134 2020-10-09 1462500 89.04 0.059997559 58.4 219.56675
## 135 2020-10-12 2573100 88.74 -2.260002136 58.4 220.16700
## 136 2020-10-13 2527500 93.00 2.919998169 58.4 220.76503
## 137 2020-10-14 4180000 86.86 -8.558998108 58.4 221.36087
## 138 2020-10-15 2029700 90.38 4.729995728 58.4 221.95454
## 139 2020-10-16 2792500 94.09 0.489997864 58.4 222.54606
## 140 2020-10-19 2800200 93.92 -2.310005188 58.4 223.13545
## 141 2020-10-20 1872100 94.35 0.430000305 58.4 223.72274
## 142 2020-10-21 3384600 87.72 -5.029998779 58.4 224.30796
## 143 2020-10-22 2205900 86.36 -2.629997253 58.4 224.89112
## 144 2020-10-23 2079700 88.69 -0.915000916 58.4 225.47224
## 145 2020-10-26 2879900 82.93 -5.290000916 58.4 226.05135
## 146 2020-10-27 2222100 83.95 0.949996948 58.4 226.62846
## 147 2020-10-28 2566100 77.54 -2.349998474 58.4 227.20360
## 148 2020-10-29 2512000 84.81 3.319999695 58.4 227.77679
## 149 2020-10-30 2253000 85.36 -1.650001526 58.4 228.34805
## 150 2020-11-02 2139900 87.10 -0.804000854 58.4 228.91739
## 151 2020-11-03 1082600 87.22 2.309997559 58.4 229.48484
## 152 2020-11-04 1701500 90.69 4.180000305 58.4 230.05041
## 153 2020-11-05 1459800 92.00 0.669998169 58.4 230.61412
## 154 2020-11-06 861300 92.00 0.000000000 58.4 231.17600
## 155 2020-11-09 14614900 104.80 -9.219993591 58.4 231.73605
## 156 2020-11-10 8029100 112.76 2.760002136 58.4 232.29430
## 157 2020-11-11 3274100 109.45 -2.550003052 58.4 232.85076
## 158 2020-11-12 3596500 101.63 -5.180000305 58.4 233.40546
## 159 2020-11-13 2997500 106.00 1.879997253 58.4 233.95840
## 160 2020-11-16 11171600 91.52 -5.120002747 58.4 234.50960
## 161 2020-11-17 4057400 86.93 -2.069999695 58.4 235.05909
## 162 2020-11-18 5772600 90.44 -0.360000610 58.4 235.60687
## 163 2020-11-19 3400100 94.93 3.330001831 58.4 236.15296
## 164 2020-11-20 6776800 104.07 4.069999695 58.4 236.69739
## 165 2020-11-23 5194200 106.50 -3.230003357 58.4 237.24015
## 166 2020-11-24 3864000 101.87 0.950004578 58.4 237.78127
## 167 2020-11-25 3007400 104.96 4.959999084 58.4 238.32076
## 168 2020-11-27 3874600 109.99 3.549995422 58.4 238.85864
## 169 2020-11-30 9934800 124.24 10.599998474 58.4 239.39493
## 170 2020-12-01 9634500 114.01 -10.489997864 58.4 239.92962
## 171 2020-12-02 7164700 121.09 0.239997864 58.4 240.46275
## 172 2020-12-03 4264600 118.68 -1.169998169 58.4 240.99432
## 173 2020-12-04 3142600 120.00 2.157997131 58.4 241.52435
## 174 2020-12-07 4925300 125.70 4.699996948 58.4 242.05285
## 175 2020-12-08 4597500 128.11 1.110000610 58.4 242.57983
## 176 2020-12-09 5265200 122.81 -6.180007935 58.4 243.10531
## 177 2020-12-10 5381300 129.54 5.719993591 58.4 243.62930
## 178 2020-12-11 4303700 127.30 -2.800003052 58.4 244.15181
## 179 2020-12-14 10907900 108.27 -18.359001160 58.4 244.67285
## 180 2020-12-15 5711800 111.20 -3.010002136 58.4 245.19244
## 181 2020-12-16 4802600 105.78 -6.440002441 58.4 245.71059
## 182 2020-12-17 3487800 106.43 3.794998169 58.4 246.22731
## 183 2020-12-18 3108200 104.24 -2.220001221 58.4 246.74261
## 184 2020-12-21 2611000 106.46 -1.050003052 58.4 247.25651
## 185 2020-12-22 5021000 100.56 -6.510002136 58.4 247.76901
## 186 2020-12-23 3414000 100.06 2.939994812 58.4 248.28013
## 187 2020-12-24 1039100 96.96 -1.540000916 58.4 248.78988
## lower95 price_change
## 1 44.4773150 -5.4800034
## 2 38.7103494 -1.4900017
## 3 34.2852010 -5.6500015
## 4 30.5546284 -5.6700020
## 5 27.2679280 -12.8000031
## 6 24.2965236 -13.4500008
## 7 21.5640353 -12.9900017
## 8 19.0206972 -16.6900024
## 9 16.6319419 -16.7700005
## 10 14.3726009 -19.8199997
## 11 12.2236741 -19.6100006
## 12 10.1704005 -17.5900002
## 13 8.2010413 -15.0300026
## 14 6.3060785 -16.1400032
## 15 4.4776684 -4.9000015
## 16 2.7092553 -8.4000015
## 17 0.9952943 -10.6500015
## 18 -0.6689549 -9.0000000
## 19 -2.2875822 -12.5699997
## 20 -3.8641456 -9.1500015
## 21 -5.4017635 -10.9200020
## 22 -6.9031869 -13.2900009
## 23 -8.3708576 -12.6200027
## 24 -9.8069543 -8.4000015
## 25 -11.2134313 -10.6200027
## 26 -12.5920489 -9.0800018
## 27 -13.9444000 -9.1599998
## 28 -15.2719309 -8.4200020
## 29 -16.5759601 -9.0699997
## 30 -17.8576934 -9.9099998
## 31 -19.1182365 -9.7400017
## 32 -20.3586071 -8.9099998
## 33 -21.5797436 1.7699966
## 34 -22.7825140 -7.8100014
## 35 -23.9677229 -5.3600006
## 36 -25.1361178 -6.5000000
## 37 -26.2883946 -7.3700027
## 38 -27.4252024 -7.5200005
## 39 -28.5471482 -10.8100014
## 40 -29.6547998 -11.5300026
## 41 -30.7486902 -8.8700027
## 42 -31.8293199 -6.1000023
## 43 -32.8971597 -6.0200005
## 44 -33.9526532 -7.2400017
## 45 -34.9962192 -8.4000015
## 46 -36.0282530 -9.2200012
## 47 -37.0491286 -9.6200027
## 48 -38.0592005 -10.6100006
## 49 -39.0588044 -10.9900017
## 50 -40.0482592 -9.8000031
## 51 -41.0278681 -9.8199997
## 52 -41.9979190 -7.5500031
## 53 -42.9586866 -8.5000000
## 54 -43.9104322 -7.5500031
## 55 -44.8534055 -7.4600029
## 56 -45.7878445 -7.0400009
## 57 -46.7139769 -0.9500008
## 58 -47.6320203 -5.6500015
## 59 -48.5421831 -3.6700020
## 60 -49.4446648 1.3099976
## 61 -50.3396566 3.7299995
## 62 -51.2273420 7.4000015
## 63 -52.1078971 8.3399963
## 64 -52.9814909 5.7399979
## 65 -53.8482860 4.8699989
## 66 -54.7084388 7.1100006
## 67 -55.5620995 10.9399948
## 68 -56.4094130 9.1599960
## 69 -57.2505189 7.2099991
## 70 -58.0855515 11.9599991
## 71 -58.9146406 19.3799973
## 72 -59.7379114 25.4799957
## 73 -60.5554846 19.5999985
## 74 -61.3674768 17.4700012
## 75 -62.1740009 26.8499985
## 76 -62.9751659 29.7999954
## 77 -63.7710772 33.1999969
## 78 -64.5618368 45.7699966
## 79 -65.3475434 30.1100006
## 80 -66.1282928 25.7699966
## 81 -66.9041775 28.1299973
## 82 -67.6752874 26.9799957
## 83 -68.4417096 24.0999985
## 84 -69.2035285 25.7600021
## 85 -69.9608261 23.7099991
## 86 -70.7136820 27.0699997
## 87 -71.4621734 24.7699966
## 88 -72.2063754 22.9000015
## 89 -72.9463608 18.6800003
## 90 -73.6822005 18.5999985
## 91 -74.4139634 15.5499954
## 92 -75.1417166 10.0499954
## 93 -75.8655253 10.7099991
## 94 -76.5854528 12.5999985
## 95 -77.3015611 10.2399979
## 96 -78.0139101 11.8600006
## 97 -78.7225586 9.4799957
## 98 -79.4275635 7.1999969
## 99 -80.1289806 7.8699951
## 100 -80.8268640 14.6199951
## 101 -81.5212667 12.4499969
## 102 -82.2122401 12.0899963
## 103 -82.8998346 9.5000000
## 104 -83.5840993 7.1899948
## 105 -84.2650822 5.5999985
## 106 -84.9428298 2.8499985
## 107 -85.6173881 -0.5900002
## 108 -86.2888014 3.5900002
## 109 -86.9571136 0.8799973
## 110 -87.6223670 0.1699982
## 111 -88.2846033 1.4300003
## 112 -88.9438633 3.9199982
## 113 -89.6001867 2.7999992
## 114 -90.2536123 6.7299957
## 115 -90.9041783 9.0800018
## 116 -91.5519218 8.5199966
## 117 -92.1968793 7.6399994
## 118 -92.8390864 8.3799973
## 119 -93.4785780 8.5999985
## 120 -94.1153882 7.7200012
## 121 -94.7495506 8.6500015
## 122 -95.3810978 5.4300003
## 123 -96.0100620 4.4300003
## 124 -96.6364746 8.2900009
## 125 -97.2603663 8.0999985
## 126 -97.8817675 8.4599991
## 127 -98.5007076 10.8300018
## 128 -99.1172157 14.3099976
## 129 -99.7313202 15.2999954
## 130 -100.3430491 22.2999954
## 131 -100.9524297 27.1899948
## 132 -101.5594887 27.8300018
## 133 -102.1642527 29.5499954
## 134 -102.7667473 30.6399994
## 135 -103.3669980 30.3399963
## 136 -103.9650296 34.5999985
## 137 -104.5608666 28.4599991
## 138 -105.1545329 31.9799957
## 139 -105.7460521 35.6899948
## 140 -106.3354474 35.5199966
## 141 -106.9227414 35.9499969
## 142 -107.5079565 29.3199997
## 143 -108.0911146 27.9599991
## 144 -108.6722372 30.2900009
## 145 -109.2513454 24.5299988
## 146 -109.8284602 25.5499954
## 147 -110.4036019 19.1399994
## 148 -110.9767907 26.4099960
## 149 -111.5480463 26.9599991
## 150 -112.1173880 28.6999969
## 151 -112.6848352 28.8199997
## 152 -113.2504064 32.2900009
## 153 -113.8141203 33.5999985
## 154 -114.3759949 33.5999985
## 155 -114.9360482 46.4000015
## 156 -115.4942978 54.3600006
## 157 -116.0507610 51.0499954
## 158 -116.6054548 43.2299957
## 159 -117.1583960 47.5999985
## 160 -117.7096011 33.1199951
## 161 -118.2590864 28.5299988
## 162 -118.8068679 32.0400009
## 163 -119.3529612 36.5299988
## 164 -119.8973820 45.6699982
## 165 -120.4401454 48.0999985
## 166 -120.9812666 43.4700012
## 167 -121.5207604 46.5599976
## 168 -122.0586413 51.5899963
## 169 -122.5949237 65.8399963
## 170 -123.1296218 55.6100006
## 171 -123.6627496 62.6899948
## 172 -124.1943209 60.2799988
## 173 -124.7243490 61.5999985
## 174 -125.2528475 67.2999954
## 175 -125.7798295 69.7099991
## 176 -126.3053080 64.4099960
## 177 -126.8292958 71.1399918
## 178 -127.3518054 68.9000015
## 179 -127.8728493 49.8699951
## 180 -128.3924399 52.7999954
## 181 -128.9105891 47.3799973
## 182 -129.4273090 48.0299988
## 183 -129.9426112 45.8399963
## 184 -130.4565074 48.0599976
## 185 -130.9690091 42.1599960
## 186 -131.4801274 41.6599960
## 187 -131.9898737 38.5599976# the estimated total stock value change between April and August 2020 as result of the pandemic
sum(post_covid_bio$price_change)## [1] 2757.56# Data viz
plot_ly() %>%
add_ribbons(x = post_covid_bio$date,
ymin = post_covid_bio$adj.close_bio,
ymax = post_covid_bio$yhat,
line = list(color = 'rgba(255, 0, 0, 0.05)'),
fillcolor = 'rgba(255, 0, 0, 0.6)',
name = "Estimated Change") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(bio$adj.close_bio),
yend = max(bio$adj.close_bio) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2020-01-01"),
y = max(bio$adj.close_bio) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2020-11-01"),
y = max(bio$adj.close_bio) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = paste("Estimated increase in", "<br>",
"stock price: ~$2,800",
sep = ""),
x = as.Date("2020-11-01"),
y = 6 * 10^1,
arrowhead = 1,
ax = -110,
ay = 40,
showarrow = TRUE) %>%
add_lines(x = bio$date,
y = bio$adj.close_bio,
line = list(color = "#1f77b4"),
name = "Actual") %>%
layout(title = "Covid19 Impact on Stock Value of BIONTECH",
yaxis = list(title = "Stock Price (US Dollars)"),
xaxis = list(title = "Time Series Model - Auto ARIMA Model",
range = c(as.Date("2019-10-01"), as.Date("2021-01-01"))),
legend = list(x = 0, y = 0.95))4. MODERNA
library(dplyr)
library(plotly)
library(ggplot2)
library(wesanderson)
# MODERNA
mod <- dt2[3:249,c(1,27,28,29)]
mod$Companies <- as.Date(mod$Companies)
mod$MODERNA.4 <- as.numeric(as.character(mod$MODERNA.4))
mod$MODERNA.5 <- as.numeric(as.character(mod$MODERNA.5))
mod$MODERNA.6 <- as.numeric(as.character(mod$MODERNA.6))
colnames(mod) <- c("date", "volume_mod", "adj.close_mod", "variation_mod")
plot_ly(data = mod,
x = ~ date,
y = ~ volume_mod,
type = "scatter",
mode = "line",
name = "Trading Volume of MODERNA") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(mod$volume_mod),
yend = max(mod$volume_mod) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(mod$volume_mod) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(mod$volume_mod) * 1.05,
showarrow = FALSE) %>%
layout(title = "Trading Volume of MODERNA Inc.",
yaxis = list(title = "Volume"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))plot_ly(data = mod,
x = ~ date,
y = ~ adj.close_mod,
type = "scatter",
mode = "line",
name = "Ajusted Close Price of MODERNA") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(mod$adj.close_mod),
yend = max(mod$adj.close_mod) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(mod$adj.close_mod) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(mod$adj.close_mod) * 1.05,
showarrow = FALSE) %>%
layout(title = "Adjusted Close Price of MODERNA Inc.",
yaxis = list(title = "Adjusted Close Price"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))mod %>%
ggplot(aes(x = date, y = variation_mod)) +
geom_line(color = wes_palette('Moonrise2', n=1, 'discrete')) +
geom_smooth(se = F, lty = 2, color = wes_palette('BottleRocket1', 1)) +
labs(x = "Date", y = "Variation",
title = "MODERNA Stock Price Variation",
subtitle = "Smoothed Line in Red") +
cowplot::theme_cowplot()## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
library(dplyr)
library(tsbox)
library(TSstudio)
## Quantify impact of Covid-19 on share prices
pre_covid_mod <- mod %>%
dplyr::filter(date < as.Date("2020-04-01")) %>%
dplyr::arrange(date)
post_covid_mod <- mod %>%
dplyr::filter(date >= as.Date("2020-04-01")) %>%
dplyr::arrange(date)
## The series before the outbreak of the pandemic
# Step 01: convert the series into a 'ts' object first.
ts.obj_mod <- ts(pre_covid_mod$adj.close_mod, start = c(2020, 3), frequency = 12)
ts_plot(ts.obj_mod,
title = "Stock Value of MODERNA",
Ytitle = "Price per share (US Dollars)",
slider = TRUE)ts_seasonal(ts.obj = ts.obj_mod, type = "all")# Step 02: review the series correlation with its past lags using the ACF and PACF to select time-series models for seasonal data functions.
ts_cor(ts.obj = ts.obj_mod, lag.max = 36)## Step 03: select time-series models for seasonal data
methods <- list(
auto_arima = list(method = "auto.arima",
notes = "Auto ARIMA"),
hw1 = list(method = "HoltWinters",
method_arg = NULL,
notes = "HoltWinters Model"),
hw2 = list(method = "HoltWinters",
method_arg = list(seasonal = "multiplicative"),
notes = "HW with multip. seasonality"),
tslm = list(method = "tslm",
method_arg = list(formula = input ~ trend + season),
notes = "tslm with trend and seasonal"))
# Backtesting is the time series equivalent of the machine learning cross-validation training approach. The idea here is simple - test each model with backtesting, and select the one that performed best, on average, on the different testing partition.
# For the backtesting, we will split the series into 6 testing partitions, each 12 months spaced by 3 months from each other.
train_method = list(partitions = 6,
sample.out = 12,
space = 3)
# train the models:
# Note that the forecast horizon is set the the length of the post_covid series. In addition we will set the MAPA as the error metric to evaluate the performance of the different models on the testing partitions:
md_mod <- train_model(input = ts.obj_mod,
methods = methods,
train_method = train_method,
horizon = nrow(post_covid_mod),
error = "MAPE")## # A tibble: 4 x 7
## model_id model notes avg_mape avg_rmse `avg_coverage_8~ `avg_coverage_9~
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
## 1 tslm tslm tslm wi~ 0.128 4.06 0.569 0.792
## 2 hw2 HoltWi~ HW with~ 0.176 5.25 0.625 0.819
## 3 auto_ari~ auto.a~ Auto AR~ 0.179 5.44 0.472 0.736
## 4 hw1 HoltWi~ HoltWin~ 0.180 5.33 0.639 0.819# Quantify the Covid-19 impact
# select the tslm with trend and seasonal (tslm) to calculate the Covid-19 effect when it demonstrated the lowest error through back-testing
post_covid_mod$yhat <- as.numeric(md_mod$forecast$tslm$forecast$mean)
post_covid_mod$upper95 <- as.numeric(md_mod$forecast$tslm$forecast$upper[,2])
post_covid_mod$lower95 <- as.numeric(md_mod$forecast$tslm$forecast$lower[,2])
post_covid_mod$price_change <-
post_covid_mod$adj.close_mod - post_covid_mod$yhat
post_covid_mod## date volume_mod adj.close_mod variation_mod yhat upper95
## 1 2020-04-01 7004600 29.670 -0.37999916 29.60756 35.59695
## 2 2020-04-02 15934400 33.200 2.60000038 29.29956 35.28895
## 3 2020-04-03 16896200 34.840 2.02999878 30.66556 36.65495
## 4 2020-04-06 10841600 34.640 -0.27000046 29.62456 35.61395
## 5 2020-04-07 10104300 32.020 -1.79000092 29.30356 35.29295
## 6 2020-04-08 4119800 32.420 0.05999756 29.37556 35.36495
## 7 2020-04-09 4861800 31.860 -0.68999863 29.39956 35.38895
## 8 2020-04-13 5603900 32.900 0.90000153 28.54356 34.53295
## 9 2020-04-14 8126900 34.660 0.65999985 28.50556 34.49495
## 10 2020-04-15 15683000 37.250 2.61000061 29.01156 35.00095
## 11 2020-04-16 20670700 40.600 2.52599716 30.26381 36.23128
## 12 2020-04-17 32286000 46.850 -2.14000320 31.58381 37.55128
## 13 2020-04-20 34102500 51.690 6.68999863 31.75007 37.86083
## 14 2020-04-21 39642000 49.260 -3.74000168 31.44207 37.55283
## 15 2020-04-22 11670100 51.200 0.59000015 32.80807 38.91883
## 16 2020-04-23 15152400 47.440 -1.65999985 31.76707 37.87783
## 17 2020-04-24 11702100 50.500 3.34999847 31.44607 37.55683
## 18 2020-04-27 13744400 48.050 -3.47999954 31.51807 37.62883
## 19 2020-04-28 13067200 47.880 -2.52000046 31.54207 37.65283
## 20 2020-04-29 11762000 46.370 -0.22999954 30.68607 36.79683
## 21 2020-04-30 7741900 45.990 -0.78999710 30.64807 36.75883
## 22 2020-05-01 21701300 47.930 -0.77999878 31.15407 37.26483
## 23 2020-05-04 10460500 50.500 2.45000076 32.40633 38.51279
## 24 2020-05-05 11160200 49.340 -1.13999939 33.72633 39.83279
## 25 2020-05-06 6692300 48.950 0.50000000 33.89259 40.15593
## 26 2020-05-07 30942700 53.190 -2.59000015 33.58459 39.84793
## 27 2020-05-08 24083200 59.250 6.18999863 34.95059 41.21393
## 28 2020-05-11 34852000 66.900 7.30000305 33.90959 40.17293
## 29 2020-05-12 25934500 62.350 -3.63999939 33.58859 39.85193
## 30 2020-05-13 22126000 65.180 1.38000107 33.66059 39.92393
## 31 2020-05-14 12198200 64.560 -0.23000336 33.68459 39.94793
## 32 2020-05-15 16341900 66.690 2.90000153 32.82859 39.09193
## 33 2020-05-18 80813300 80.000 -6.13999939 32.79059 39.05393
## 34 2020-05-19 64932300 71.670 -6.40000153 33.29659 39.55993
## 35 2020-05-20 53655300 73.470 3.45999908 34.54885 40.82473
## 36 2020-05-21 35318000 67.050 -6.00000000 35.86885 42.14473
## 37 2020-05-22 28912700 69.000 -0.93000031 36.03511 42.48003
## 38 2020-05-26 40332600 57.710 -8.69000244 35.72711 42.17203
## 39 2020-05-27 59514700 52.180 0.97999954 37.09311 43.53803
## 40 2020-05-28 32258900 55.540 2.04000092 36.05211 42.49703
## 41 2020-05-29 38070400 61.500 4.04999924 35.73111 42.17603
## 42 2020-06-01 21257500 62.180 -0.57999802 35.80311 42.24803
## 43 2020-06-02 21701100 59.870 1.16999817 35.82711 42.27203
## 44 2020-06-03 20811500 59.890 2.38000107 34.97111 41.41603
## 45 2020-06-04 13504000 60.590 -1.02000046 34.93311 41.37803
## 46 2020-06-05 14677900 58.190 -0.30000305 35.43911 41.88403
## 47 2020-06-08 11249300 59.100 0.79999924 36.69137 43.16471
## 48 2020-06-09 7600200 58.230 -0.31999969 38.01137 44.48471
## 49 2020-06-10 11074400 60.070 1.27000046 38.17763 44.83076
## 50 2020-06-11 32085200 60.200 -2.75000000 37.86963 44.52276
## 51 2020-06-12 14742900 62.000 -0.35499954 39.23563 45.88876
## 52 2020-06-15 19339300 66.570 0.66999817 38.19463 44.84776
## 53 2020-06-16 16537800 64.340 -1.97000122 37.87363 44.52676
## 54 2020-06-17 13616000 63.310 -0.28999710 37.94563 44.59876
## 55 2020-06-18 10914400 64.950 1.91999817 37.96963 44.62276
## 56 2020-06-19 16076700 66.350 1.36000061 37.11363 43.76676
## 57 2020-06-22 13586400 64.750 -1.16999817 37.07563 43.72876
## 58 2020-06-23 12361300 62.940 -1.15999985 37.58163 44.23476
## 59 2020-06-24 14377100 64.840 1.60999680 38.83389 45.53023
## 60 2020-06-25 26905700 62.000 -5.00000000 40.15389 46.85023
## 61 2020-06-26 13866300 61.280 0.56999969 40.32015 47.20570
## 62 2020-06-29 8074200 62.080 0.26000214 40.01215 46.89770
## 63 2020-06-30 11955200 64.210 2.02999878 41.37815 48.26370
## 64 2020-07-01 17916500 61.590 -1.40999985 40.33715 47.22270
## 65 2020-07-02 26544400 58.570 -1.43000031 40.01615 46.90170
## 66 2020-07-06 9977300 59.150 0.60000229 40.08815 46.97370
## 67 2020-07-07 8904300 61.090 2.29000092 40.11215 46.99770
## 68 2020-07-08 7590900 61.580 0.79000092 39.25615 46.14170
## 69 2020-07-09 20646100 64.960 2.86999893 39.21815 46.10370
## 70 2020-07-10 7407500 62.610 -2.04000092 39.72415 46.60970
## 71 2020-07-13 56617300 71.780 7.88999939 40.97641 47.91885
## 72 2020-07-14 38708800 75.040 1.04000092 42.29641 49.23885
## 73 2020-07-15 90756100 80.220 -7.29999542 42.46267 49.60250
## 74 2020-07-16 27277400 81.810 1.64999390 42.15467 49.29450
## 75 2020-07-17 103509200 94.850 11.50000000 43.52067 50.66050
## 76 2020-07-20 78391300 82.680 -5.18000031 42.47967 49.61950
## 77 2020-07-21 26157900 80.860 -0.05000305 42.15867 49.29850
## 78 2020-07-22 27274000 83.225 3.61499786 42.23067 49.37050
## 79 2020-07-23 29288100 75.330 -6.72000122 42.25467 49.39450
## 80 2020-07-24 27797500 73.210 4.15999603 41.39867 48.53850
## 81 2020-07-27 44710700 79.910 -0.71999359 41.36067 48.50050
## 82 2020-07-28 32946900 81.490 0.48999786 41.86667 49.00650
## 83 2020-07-29 24453900 79.510 -2.47000122 43.11893 50.32820
## 84 2020-07-30 13102000 77.630 -0.37000275 44.43893 51.64820
## 85 2020-07-31 14210400 74.100 -4.09999847 44.60519 52.01889
## 86 2020-08-03 13382300 77.980 3.00000000 44.29719 51.71089
## 87 2020-08-04 10126500 78.460 0.26799774 45.66319 53.07689
## 88 2020-08-05 19831900 75.800 -1.68999481 44.62219 52.03589
## 89 2020-08-06 11691500 73.760 -2.02999878 44.30119 51.71489
## 90 2020-08-07 8771400 74.100 -0.27999878 44.37319 51.78689
## 91 2020-08-10 8678300 72.010 -2.25999451 44.39719 51.81089
## 92 2020-08-11 18118700 68.970 -0.54999542 43.54119 50.95489
## 93 2020-08-12 50362100 69.520 -6.15000153 43.50319 50.91689
## 94 2020-08-13 15055900 67.830 -1.25999451 44.00919 51.42289
## 95 2020-08-14 12799600 69.150 1.09000397 45.26144 52.75606
## 96 2020-08-17 8749100 69.760 0.62000275 46.58144 54.07606
## 97 2020-08-18 11329200 67.030 -1.25000000 46.74770 54.45280
## 98 2020-08-19 8558800 68.000 1.26000214 46.43970 54.14480
## 99 2020-08-20 6346700 67.810 0.01999664 47.80570 55.51080
## 100 2020-08-21 6935600 66.450 -1.05000305 46.76470 54.46980
## 101 2020-08-24 13143200 64.880 -2.27000427 46.44370 54.14880
## 102 2020-08-25 7798500 66.250 1.93000031 46.51570 54.22080
## 103 2020-08-26 28768100 70.500 2.87999725 46.53970 54.24480
## 104 2020-08-27 9262600 68.030 -1.95000458 45.68370 53.38880
## 105 2020-08-28 6368200 67.490 -0.55000305 45.64570 53.35080
## 106 2020-08-31 15271800 64.890 -2.25000000 46.15170 53.85680
## 107 2020-09-01 8183600 63.320 -0.22000122 47.40396 55.20041
## 108 2020-09-02 9874600 64.720 1.88000107 48.72396 56.52041
## 109 2020-09-03 14307300 64.840 -0.81000519 48.89022 56.90231
## 110 2020-09-04 11653300 62.600 -2.38999939 48.58222 56.59431
## 111 2020-09-08 17706200 54.340 -3.40999985 49.94822 57.96031
## 112 2020-09-09 9355400 56.900 0.87000275 48.90722 56.91931
## 113 2020-09-10 7985300 57.560 0.51900101 48.58622 56.59831
## 114 2020-09-11 4821900 59.340 0.84999847 48.65822 56.67031
## 115 2020-09-14 8930800 63.670 3.51999664 48.68222 56.69431
## 116 2020-09-15 9101200 66.880 1.85299683 47.82622 55.83831
## 117 2020-09-16 11345200 68.840 1.19999695 47.78822 55.80031
## 118 2020-09-17 13275200 67.890 -0.61000061 48.29422 56.30631
## 119 2020-09-18 13206200 69.870 1.59000397 49.54648 57.65939
## 120 2020-09-21 8223000 69.260 0.66000366 50.86648 58.97939
## 121 2020-09-22 6303600 68.720 1.22000122 51.03274 59.36569
## 122 2020-09-23 6919400 67.160 -0.70499420 50.72474 59.05769
## 123 2020-09-24 5565200 65.170 -0.33000183 52.09074 60.42369
## 124 2020-09-25 9190600 69.470 3.58300018 51.04974 59.38269
## 125 2020-09-28 7337400 70.550 0.96000671 50.72874 59.06169
## 126 2020-09-29 6027600 70.520 0.05999756 50.80074 59.13369
## 127 2020-09-30 15690700 70.750 -1.76000214 50.82474 59.15769
## 128 2020-10-01 6839000 70.030 0.45999908 49.96874 58.30169
## 129 2020-10-02 8004900 68.810 0.90999603 49.93074 58.26369
## 130 2020-10-05 6518100 71.950 2.30999756 50.43674 58.76969
## 131 2020-10-06 7285500 71.040 -0.95999908 51.68900 60.13137
## 132 2020-10-07 4190600 72.370 1.13999939 53.00900 61.45137
## 133 2020-10-08 4023800 72.930 -0.56999969 53.17526 61.84141
## 134 2020-10-09 4774000 73.000 -0.41999817 52.86726 61.53341
## 135 2020-10-12 7590800 75.310 1.69999695 54.23326 62.89941
## 136 2020-10-13 13669200 78.290 2.36000061 53.19226 61.85841
## 137 2020-10-14 6634600 76.560 -3.45000458 52.87126 61.53741
## 138 2020-10-15 5009100 75.580 0.09999847 52.94326 61.60941
## 139 2020-10-16 4693600 73.940 -2.11999512 52.96726 61.63341
## 140 2020-10-19 6931300 70.960 -3.01000214 52.11126 60.77741
## 141 2020-10-20 5441300 71.310 -0.76000214 52.07326 60.73941
## 142 2020-10-21 4614800 68.370 -3.19999695 52.57926 61.24541
## 143 2020-10-22 5212100 70.840 2.33599854 53.83152 62.61489
## 144 2020-10-23 3735100 70.530 -1.77000427 55.15152 63.93489
## 145 2020-10-26 3872800 70.240 -0.12000275 55.31778 64.32810
## 146 2020-10-27 3812900 70.670 0.22000122 55.00978 64.02010
## 147 2020-10-28 6044000 65.740 -3.20999908 56.37578 65.38610
## 148 2020-10-29 13776400 71.280 2.93000031 55.33478 64.34510
## 149 2020-10-30 7070500 67.470 -3.33000183 55.01378 64.02410
## 150 2020-11-02 4650300 67.110 -0.90999603 55.08578 64.09610
## 151 2020-11-03 4366900 69.080 2.33000183 55.10978 64.12010
## 152 2020-11-04 5442100 69.810 0.83999634 54.25378 63.26410
## 153 2020-11-05 4449900 71.480 0.39000702 54.21578 63.22610
## 154 2020-11-06 5841000 72.450 1.02999878 54.72178 63.73210
## 155 2020-11-09 20266800 77.740 -0.01000214 55.97404 65.10864
## 156 2020-11-10 10499400 76.050 -3.94999695 57.29404 66.42864
## 157 2020-11-11 18576500 82.440 5.38999939 57.46030 66.82454
## 158 2020-11-12 21700100 87.810 2.62999725 57.15230 66.51655
## 159 2020-11-13 14121100 89.390 2.76999664 58.51830 67.88254
## 160 2020-11-16 70853100 97.950 -2.36000061 57.47730 66.84154
## 161 2020-11-17 26172000 93.150 -4.08000183 57.15630 66.52054
## 162 2020-11-18 21823500 88.890 -4.98000336 57.22830 66.59254
## 163 2020-11-19 16002000 92.770 2.76999664 57.25230 66.61654
## 164 2020-11-20 15306300 97.610 2.61000061 56.39630 65.76054
## 165 2020-11-23 16359600 101.030 0.26000214 56.35830 65.72254
## 166 2020-11-24 12515300 98.560 0.15999603 56.86430 66.22854
## 167 2020-11-25 25850700 109.180 10.15000153 58.11656 67.61151
## 168 2020-11-27 43438500 127.030 11.90999603 59.43656 68.93151
## 169 2020-11-30 73203100 152.740 8.74000549 59.60281 69.32968
## 170 2020-12-01 125552300 141.010 -36.65000916 59.29482 69.02168
## 171 2020-12-02 51217300 143.000 4.36999512 60.66082 70.38768
## 172 2020-12-03 38656500 157.260 13.75999451 59.61982 69.34668
## 173 2020-12-04 27704800 152.520 -5.06999207 59.29882 69.02568
## 174 2020-12-07 31409800 159.520 4.49000549 59.37082 69.09768
## 175 2020-12-08 27738700 169.860 4.11000061 59.39482 69.12168
## 176 2020-12-09 40951700 156.590 -13.61000061 58.53882 68.26568
## 177 2020-12-10 22060000 155.690 4.55999756 58.50082 68.22768
## 178 2020-12-11 22069900 156.930 -2.77000427 59.00682 68.73368
## 179 2020-12-14 22467700 155.070 -8.32998657 60.25907 70.12248
## 180 2020-12-15 36451300 147.220 -9.77999878 61.57907 71.44248
## 181 2020-12-16 28476300 137.030 -0.72000122 61.74533 71.84258
## 182 2020-12-17 32404500 144.000 6.08000183 61.43733 71.53458
## 183 2020-12-18 33157800 140.230 -1.27999878 62.80333 72.90058
## 184 2020-12-21 23921000 138.300 -2.53999329 61.76233 71.85958
## 185 2020-12-22 36035800 125.880 -13.12000275 61.44133 71.53858
## lower95 price_change
## 1 23.61816 0.06244488
## 2 23.31016 3.90044460
## 3 24.67616 4.17444477
## 4 23.63516 5.01544361
## 5 23.31416 2.71644497
## 6 23.38616 3.04444256
## 7 23.41016 2.46044483
## 8 22.55416 4.35644588
## 9 22.51616 6.15444393
## 10 23.02216 8.23844433
## 11 24.29635 10.33618383
## 12 25.61635 15.26618382
## 13 25.63932 19.93992488
## 14 25.33132 17.81792361
## 15 26.69732 18.39192683
## 16 25.65632 15.67292430
## 17 25.33532 19.05392596
## 18 25.40732 16.53192507
## 19 25.43132 16.33792674
## 20 24.57532 15.68392474
## 21 24.53732 15.34192721
## 22 25.04332 16.77592608
## 23 26.29987 18.09366681
## 24 27.61987 15.61366695
## 25 27.62926 15.05740846
## 26 27.32126 19.60540536
## 27 28.68726 24.29940751
## 28 27.64626 32.99040864
## 29 27.32526 28.76140588
## 30 27.39726 31.51940759
## 31 27.42126 30.87540468
## 32 26.56526 33.86140970
## 33 26.52726 47.20940698
## 34 27.03326 38.37340539
## 35 28.27297 38.92114948
## 36 29.59297 31.18115129
## 37 29.59019 32.96488914
## 38 29.28219 21.98288727
## 39 30.64819 15.08688927
## 40 29.60719 19.48788948
## 41 29.28619 25.76888885
## 42 29.35819 26.37688904
## 43 29.38219 24.04288750
## 44 28.52619 24.91888809
## 45 28.48819 25.65688858
## 46 28.99419 22.75088730
## 47 30.21803 22.40862818
## 48 31.53803 20.21862923
## 49 31.52450 21.89237029
## 50 31.21650 22.33037039
## 51 32.58250 22.76437041
## 52 31.54150 28.37536971
## 53 31.22050 26.46636664
## 54 31.29250 25.36437155
## 55 31.31650 26.98036696
## 56 30.46050 29.23636862
## 57 30.42250 27.67436988
## 58 30.92850 25.35836875
## 59 32.13754 26.00610749
## 60 33.45754 21.84611114
## 61 33.43459 20.95985082
## 62 33.12659 22.06785291
## 63 34.49259 22.83185094
## 64 33.45159 21.25285161
## 65 33.13059 18.55385145
## 66 33.20259 19.06185315
## 67 33.22659 20.97785161
## 68 32.37059 22.32385343
## 69 32.33259 25.74185041
## 70 32.83859 22.88585218
## 71 34.03396 30.80359138
## 72 35.35396 32.74359350
## 73 35.32284 37.75733471
## 74 35.01484 39.65533009
## 75 36.38084 51.32933178
## 76 35.33984 40.20033321
## 77 35.01884 38.70133381
## 78 35.09084 40.99433155
## 79 35.11484 33.07533474
## 80 34.25884 31.81133213
## 81 34.22084 38.54933643
## 82 34.72684 39.62333088
## 83 35.90965 36.39107618
## 84 37.22965 33.19107129
## 85 37.19148 29.49481341
## 86 36.88348 33.68281733
## 87 38.24948 32.79681384
## 88 37.20848 31.17781741
## 89 36.88748 29.45881678
## 90 36.95948 29.72681300
## 91 36.98348 27.61281649
## 92 36.12748 25.42881571
## 93 36.08948 26.01681086
## 94 36.59548 23.82081629
## 95 37.76683 23.88855702
## 96 39.08683 23.17855762
## 97 39.04261 20.28229516
## 98 38.73461 21.56029542
## 99 40.10061 20.00429376
## 100 39.05961 19.68529275
## 101 38.73861 18.43629335
## 102 38.81061 19.73429597
## 103 38.83461 23.96029580
## 104 37.97861 22.34629472
## 105 37.94061 21.84429353
## 106 38.44661 18.73829530
## 107 39.60752 15.91603664
## 108 40.92752 15.99603815
## 109 40.87814 15.94977417
## 110 40.57014 14.01777534
## 111 41.93614 4.39177780
## 112 40.89514 7.99277878
## 113 40.57414 8.97377892
## 114 40.64614 10.68177757
## 115 40.67014 14.98777542
## 116 39.81414 19.05377464
## 117 39.77614 21.05177345
## 118 40.28214 19.59577675
## 119 41.43357 20.32352114
## 120 42.75357 18.39352051
## 121 42.69979 17.68726050
## 122 42.39179 16.43526198
## 123 43.75779 13.07925727
## 124 42.71679 18.42025992
## 125 42.39579 19.82126204
## 126 42.46779 19.71925551
## 127 42.49179 19.92525870
## 128 41.63579 20.06125762
## 129 41.59779 18.87925612
## 130 42.10379 21.51325576
## 131 43.24663 19.35100075
## 132 44.56663 19.36100257
## 133 44.50911 19.75474103
## 134 44.20111 20.13273977
## 135 45.56711 21.07673810
## 136 44.52611 25.09774106
## 137 44.20511 23.68873800
## 138 44.27711 22.63674215
## 139 44.30111 20.97274259
## 140 43.44511 18.84873937
## 141 43.40711 19.23673757
## 142 43.91311 15.79074300
## 143 45.04815 17.00847762
## 144 46.36815 15.37848005
## 145 46.30746 14.92222004
## 146 45.99946 15.66021938
## 147 47.36546 9.36421986
## 148 46.32446 15.94522038
## 149 46.00346 12.45622311
## 150 46.07546 12.02422237
## 151 46.09946 13.97022343
## 152 45.24346 15.55621929
## 153 45.20546 17.26422482
## 154 45.71146 17.72821865
## 155 46.83943 21.76596060
## 156 48.15943 18.75596577
## 157 48.09605 24.97970606
## 158 47.78805 30.65770022
## 159 49.15405 30.87170283
## 160 48.11305 40.47269999
## 161 47.79205 35.99370486
## 162 47.86405 31.66170260
## 163 47.88805 35.51769969
## 164 47.03205 41.21370379
## 165 46.99405 44.67170169
## 166 47.50005 41.69570071
## 167 48.62161 51.06344449
## 168 49.94161 67.59344295
## 169 49.87595 93.13719056
## 170 49.56795 81.71517862
## 171 50.93395 82.33918489
## 172 49.89295 97.64017900
## 173 49.57195 93.22118905
## 174 49.64395 100.14918893
## 175 49.66795 110.46518510
## 176 48.81195 98.05118097
## 177 48.77395 97.18918680
## 178 49.27995 97.92317727
## 179 50.39567 94.81093295
## 180 51.71567 85.64092684
## 181 51.64809 75.28466530
## 182 51.34009 82.56266556
## 183 52.70609 77.42666206
## 184 51.66509 76.53766899
## 185 51.34409 64.43866348# the estimated total stock value change between April and August 2020 as result of the pandemic
sum(post_covid_mod$price_change)## [1] 5355.499# Data viz
plot_ly() %>%
add_ribbons(x = post_covid_mod$date,
ymin = post_covid_mod$adj.close_mod,
ymax = post_covid_mod$yhat,
line = list(color = 'rgba(255, 0, 0, 0.05)'),
fillcolor = 'rgba(255, 0, 0, 0.6)',
name = "Estimated Change") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(mod$adj.close_mod),
yend = max(mod$adj.close_mod) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2020-01-01"),
y = max(mod$adj.close_mod) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2020-11-01"),
y = max(mod$adj.close_mod) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = paste("Estimated increase in", "<br>",
"stock price: ~$5,400",
sep = ""),
x = as.Date("2020-09-01"),
y = 5 * 10^1,
arrowhead = 1,
ax = 120,
ay = 40,
showarrow = TRUE) %>%
add_lines(x = mod$date,
y = mod$adj.close_mod,
line = list(color = "#1f77b4"),
name = "Actual") %>%
layout(title = "Covid19 Impact on Stock Value of MODERNA",
yaxis = list(title = "Stock Price (US Dollars)"),
xaxis = list(title = "Time Series Model - Auto ARIMA Model",
range = c(as.Date("2019-10-01"), as.Date("2021-01-01"))),
legend = list(x = 0, y = 0.95))5. NOVAVAX
library(dplyr)
library(plotly)
library(ggplot2)
library(wesanderson)
# NOVAVAX
nova <- dt2[3:247,c(1,34,35,36)]
nova$Companies <- as.Date(nova$Companies)
nova$NOVAVAX.4 <- as.numeric(as.character(nova$NOVAVAX.4))
nova$NOVAVAX.5 <- as.numeric(as.character(nova$NOVAVAX.5))
nova$NOVAVAX.6 <- as.numeric(as.character(nova$NOVAVAX.6))
colnames(nova) <- c("date", "volume_nova", "adj.close_nova", "variation_nova")
plot_ly(data = nova,
x = ~ date,
y = ~ volume_nova,
type = "scatter",
mode = "line",
name = "Trading Volume of NOVAVAX") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(nova$volume_nova),
yend = max(nova$volume_nova) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(nova$volume_nova) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(nova$volume_nova) * 1.05,
showarrow = FALSE) %>%
layout(title = "Trading Volume of NOVAVAX Inc.",
yaxis = list(title = "Volume"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))plot_ly(data = nova,
x = ~ date,
y = ~ adj.close_nova,
type = "scatter",
mode = "line",
name = "Ajusted Close Price of NOVAVAX") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(nova$adj.close_nova),
yend = max(nova$adj.close_nova) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2019-09-01"),
y = max(nova$adj.close_nova) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2021-08-01"),
y = max(nova$adj.close_nova) * 1.05,
showarrow = FALSE) %>%
layout(title = "Adjusted Close Price of NOVAVAX Inc.",
yaxis = list(title = "Adjusted Close Price"),
xaxis = list(title = "Source: COVID-19 biotech companies on stock exchange(2020)"))nova %>%
ggplot(aes(x = date, y = variation_nova)) +
geom_line(color = wes_palette('Moonrise2', n=1, 'discrete')) +
geom_smooth(se = F, lty = 2, color = wes_palette('BottleRocket1', 1)) +
labs(x = "Date", y = "Variation",
title = "NOVAVAX Stock Price Variation",
subtitle = "Smoothed Line in Red") +
cowplot::theme_cowplot()## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
library(dplyr)
library(tsbox)
library(TSstudio)
## Quantify impact of Covid-19 on share prices
pre_covid_nova <- nova %>%
dplyr::filter(date < as.Date("2020-04-01")) %>%
dplyr::arrange(date)
post_covid_nova <- nova %>%
dplyr::filter(date >= as.Date("2020-04-01")) %>%
dplyr::arrange(date)
## The series before the outbreak of the pandemic
# Step 01: convert the series into a 'ts' object first.
ts.obj_nova <- ts(pre_covid_nova$adj.close_nova, start = c(2020, 3), frequency = 12)
ts_plot(ts.obj_nova,
title = "Stock Value of NOVAVAX",
Ytitle = "Price per share (US Dollars)",
slider = TRUE)ts_seasonal(ts.obj = ts.obj_nova, type = "all")# Step 02: review the series correlation with its past lags using the ACF and PACF to select time-series models for seasonal data functions.
ts_cor(ts.obj = ts.obj_nova, lag.max = 36)## Step 03: select time-series models for seasonal data
methods <- list(
auto_arima = list(method = "auto.arima",
notes = "Auto ARIMA"),
hw1 = list(method = "HoltWinters",
method_arg = NULL,
notes = "HoltWinters Model"),
hw2 = list(method = "HoltWinters",
method_arg = list(seasonal = "multiplicative"),
notes = "HW with multip. seasonality"),
tslm = list(method = "tslm",
method_arg = list(formula = input ~ trend + season),
notes = "tslm with trend and seasonal"))
# Backtesting is the time series equivalent of the machine learning cross-validation training approach. The idea here is simple - test each model with backtesting, and select the one that performed best, on average, on the different testing partition.
# For the backtesting, we will split the series into 6 testing partitions, each 12 months spaced by 3 months from each other.
train_method = list(partitions = 6,
sample.out = 12,
space = 3)
# train the models:
# Note that the forecast horizon is set the the length of the post_covid series. In addition we will set the MAPA as the error metric to evaluate the performance of the different models on the testing partitions:
md_nova <- train_model(input = ts.obj_nova,
methods = methods,
train_method = train_method,
horizon = nrow(post_covid_nova),
error = "MAPE")## # A tibble: 4 x 7
## model_id model notes avg_mape avg_rmse `avg_coverage_8~ `avg_coverage_9~
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
## 1 tslm tslm tslm wi~ 0.213 2.68 0.556 0.819
## 2 auto_ari~ auto.a~ Auto AR~ 0.236 2.97 0.764 0.917
## 3 hw1 HoltWi~ HoltWin~ 0.315 3.78 0.625 0.806
## 4 hw2 HoltWi~ HW with~ 0.380 4.60 0.736 0.861# Quantify the Covid-19 impact
# select the tslm with trend and seasonal (tslm) to calculate the Covid-19 effect when it demonstrated the lowest error through back-testing
post_covid_nova$yhat <- as.numeric(md_nova$forecast$tslm$forecast$mean)
post_covid_nova$upper95 <- as.numeric(md_nova$forecast$tslm$forecast$upper[,2])
post_covid_nova$lower95 <- as.numeric(md_nova$forecast$tslm$forecast$lower[,2])
post_covid_nova$price_change <-
post_covid_nova$adj.close_nova - post_covid_nova$yhat
post_covid_nova## date volume_nova adj.close_nova variation_nova yhat upper95
## 1 2020-04-01 5953000 14.03 -0.15000057 12.64600 16.43004
## 2 2020-04-02 3110300 13.86 0.21000004 14.06800 17.85204
## 3 2020-04-03 9144800 15.61 1.86999989 13.60400 17.38804
## 4 2020-04-06 5415800 16.10 0.50000000 13.31800 17.10204
## 5 2020-04-07 4704500 14.90 -0.89000034 13.07600 16.86004
## 6 2020-04-08 13825700 17.18 0.03000069 13.56600 17.35004
## 7 2020-04-09 4766100 17.05 -0.40000153 14.01400 17.79804
## 8 2020-04-13 4114700 17.64 -0.21000099 13.23200 17.01604
## 9 2020-04-14 3936500 18.00 0.14999962 13.63200 17.41604
## 10 2020-04-15 3455100 17.53 -0.01999855 13.93800 17.72204
## 11 2020-04-16 4188800 18.10 0.50000000 14.67567 18.44585
## 12 2020-04-17 7478600 19.08 0.97999954 14.06900 17.83918
## 13 2020-04-20 27006300 23.80 4.69999886 14.33000 18.19072
## 14 2020-04-21 15517500 21.53 -2.51000023 15.75200 19.61272
## 15 2020-04-22 6488700 21.37 -1.60999870 15.28800 19.14872
## 16 2020-04-23 4911400 20.38 -1.04000092 15.00200 18.86272
## 17 2020-04-24 4969800 19.83 -0.48999977 14.76000 18.62072
## 18 2020-04-27 5280400 21.01 0.80999947 15.25000 19.11072
## 19 2020-04-28 3983000 20.55 -0.85000038 15.69800 19.55872
## 20 2020-04-29 6512300 18.41 -1.61000061 14.91600 18.77672
## 21 2020-04-30 4548900 18.13 -0.72000122 15.31600 19.17672
## 22 2020-05-01 4761400 17.33 -0.46999931 15.62200 19.48272
## 23 2020-05-04 3736700 18.06 1.05999947 16.35967 20.21767
## 24 2020-05-05 2948300 17.78 -0.46999931 15.75300 19.61100
## 25 2020-05-06 2481000 17.14 -0.57999992 16.01400 19.97111
## 26 2020-05-07 3392200 18.15 0.84000015 17.43600 21.39311
## 27 2020-05-08 3075400 18.73 0.52999878 16.97200 20.92911
## 28 2020-05-11 24642100 24.50 4.69000053 16.68600 20.64311
## 29 2020-05-12 74649600 39.82 0.79999924 16.44400 20.40111
## 30 2020-05-13 24237100 40.67 1.34999847 16.93400 20.89111
## 31 2020-05-14 7832800 38.30 -0.93000031 17.38200 21.33911
## 32 2020-05-15 21892500 43.63 5.45000076 16.60000 20.55711
## 33 2020-05-18 24660100 56.96 7.68000031 17.00000 20.95711
## 34 2020-05-19 12579400 53.90 -4.34999847 17.30600 21.26311
## 35 2020-05-20 8180300 52.81 -1.19999695 18.04367 22.00871
## 36 2020-05-21 5479500 50.07 -2.68000031 17.43700 21.40204
## 37 2020-05-22 6083300 46.11 -2.38999939 17.69800 21.76984
## 38 2020-05-26 22045300 48.17 -6.18000031 19.12000 23.19184
## 39 2020-05-27 10439500 45.47 -2.72999954 18.65600 22.72784
## 40 2020-05-28 4955100 44.84 -1.25999832 18.37000 22.44184
## 41 2020-05-29 4119800 46.04 1.31999969 18.12800 22.19984
## 42 2020-06-01 9686800 52.82 6.65000153 18.61800 22.68984
## 43 2020-06-02 6538800 49.64 -3.10000229 19.06600 23.13784
## 44 2020-06-03 16950300 44.25 -4.65000153 18.28400 22.35584
## 45 2020-06-04 5713100 44.66 0.65999985 18.68400 22.75584
## 46 2020-06-05 8077200 46.30 -3.15999985 18.99000 23.06184
## 47 2020-06-08 4812700 44.79 0.20999908 19.72767 23.81746
## 48 2020-06-09 3875400 45.93 1.88000107 19.12100 23.21079
## 49 2020-06-10 2308700 44.96 -1.20999908 19.38200 23.58538
## 50 2020-06-11 4089100 44.43 1.15000153 20.80400 25.00738
## 51 2020-06-12 3297800 45.57 0.02000046 20.34000 24.54338
## 52 2020-06-15 8694400 51.07 5.13000107 20.05400 24.25738
## 53 2020-06-16 5026200 52.15 0.97000122 19.81200 24.01538
## 54 2020-06-17 9350000 55.41 3.38000107 20.30200 24.50538
## 55 2020-06-18 5654100 59.27 4.02000046 20.75000 24.95338
## 56 2020-06-19 11569300 64.75 1.47000122 19.96800 24.17138
## 57 2020-06-22 8431300 65.95 -1.65000153 20.36800 24.57138
## 58 2020-06-23 6138100 69.66 6.72000504 20.67400 24.87738
## 59 2020-06-24 10960400 77.50 7.44999695 21.41167 25.64235
## 60 2020-06-25 10743500 83.54 2.23000336 20.80500 25.03569
## 61 2020-06-26 13522400 77.39 -2.34000397 21.06600 25.41623
## 62 2020-06-29 8693300 83.61 2.12999725 22.48800 26.83823
## 63 2020-06-30 7782200 83.35 -1.65000153 22.02400 26.37423
## 64 2020-07-01 6612100 79.70 0.22999573 21.73800 26.08823
## 65 2020-07-02 5210900 81.64 4.16999817 21.49600 25.84623
## 66 2020-07-06 3832000 79.44 -0.70999908 21.98600 26.33623
## 67 2020-07-07 26477900 104.56 0.55999756 22.43400 26.78423
## 68 2020-07-08 7898700 98.30 -5.54999542 21.65200 26.00223
## 69 2020-07-09 4348600 96.30 -2.01999664 22.05200 26.40223
## 70 2020-07-10 3072200 94.36 0.36000061 22.35800 26.70823
## 71 2020-07-13 8636400 104.32 8.98999786 23.09567 27.48184
## 72 2020-07-14 5390800 109.97 2.76000214 22.48900 26.87517
## 73 2020-07-15 5250900 111.15 0.84000397 22.75000 27.26088
## 74 2020-07-16 6940400 120.29 10.88999939 24.17200 28.68288
## 75 2020-07-17 15200800 140.49 11.60000610 23.70800 28.21888
## 76 2020-07-20 12934500 138.23 -11.26000977 23.42200 27.93288
## 77 2020-07-21 8868700 140.69 0.18000793 23.18000 27.69088
## 78 2020-07-22 6143100 146.45 10.44999695 23.67000 28.18088
## 79 2020-07-23 5923200 139.59 -5.10000610 24.11800 28.62888
## 80 2020-07-24 4819300 133.93 0.72999573 23.33600 27.84688
## 81 2020-07-27 5035600 139.60 5.58000183 23.73600 28.24688
## 82 2020-07-28 5907800 148.60 9.17001343 24.04200 28.55288
## 83 2020-07-29 5638600 145.36 -7.13999939 24.77967 29.33442
## 84 2020-07-30 5410600 146.62 6.11999512 24.17300 28.72775
## 85 2020-07-31 4232100 143.10 -4.50999451 24.43400 29.11791
## 86 2020-08-03 6762900 155.87 10.58999634 25.85600 30.53991
## 87 2020-08-04 11517400 157.17 0.66999817 25.39200 30.07591
## 88 2020-08-05 17875600 173.49 -8.50999451 25.10600 29.78991
## 89 2020-08-06 7882700 167.50 -10.50000000 24.86400 29.54791
## 90 2020-08-07 6540200 170.29 1.65998840 25.35400 30.03791
## 91 2020-08-10 7607000 178.51 3.04998779 25.80200 30.48591
## 92 2020-08-11 15226000 149.48 -20.29000854 25.02000 29.70391
## 93 2020-08-12 17941200 124.42 -23.63999939 25.42000 30.10391
## 94 2020-08-13 10776300 133.28 7.11999512 25.72600 30.40991
## 95 2020-08-14 13721100 146.51 2.50999451 26.46367 31.19869
## 96 2020-08-17 5690500 155.53 6.52999878 25.85700 30.59203
## 97 2020-08-18 7441000 146.23 -6.44999695 26.11800 30.98600
## 98 2020-08-19 5328700 147.06 5.47999573 27.54000 32.40800
## 99 2020-08-20 3727400 143.13 0.20001221 27.07600 31.94400
## 100 2020-08-21 5660900 137.62 -4.76000977 26.79000 31.65800
## 101 2020-08-24 8483700 119.26 -12.29999542 26.54800 31.41600
## 102 2020-08-25 13008400 113.11 5.09999847 27.03800 31.90600
## 103 2020-08-26 6988800 114.62 7.62000275 27.48600 32.35400
## 104 2020-08-27 4827000 107.35 -7.95000458 26.70400 31.57200
## 105 2020-08-28 3217200 107.74 -0.36000061 27.10400 31.97200
## 106 2020-08-31 6236200 110.34 -5.16000366 27.41000 32.27800
## 107 2020-09-01 4830700 105.08 -3.13999939 28.14767 33.07338
## 108 2020-09-02 6355200 102.90 3.34000397 27.54100 32.46672
## 109 2020-09-03 7538300 102.98 -9.82999420 27.80200 32.86396
## 110 2020-09-04 7674000 92.93 -8.01000214 29.22400 34.28596
## 111 2020-09-08 5625700 85.31 -1.09000397 28.76000 33.82196
## 112 2020-09-09 3780500 90.84 2.62999725 28.47400 33.53596
## 113 2020-09-10 4656600 92.86 0.02999878 28.23200 33.29396
## 114 2020-09-11 5303400 95.45 1.44999695 28.72200 33.78396
## 115 2020-09-14 7979000 105.80 9.42000580 29.17000 34.23196
## 116 2020-09-15 6779500 110.34 0.33999634 28.38800 33.44996
## 117 2020-09-16 10805500 120.13 12.04000092 28.78800 33.84996
## 118 2020-09-17 5531300 115.51 -3.48999786 29.09400 34.15596
## 119 2020-09-18 5820800 108.36 -6.58000183 29.83167 34.95732
## 120 2020-09-21 3241200 110.15 7.65000153 29.22500 34.35066
## 121 2020-09-22 2407200 111.63 1.87999725 29.48600 34.75068
## 122 2020-09-23 3216800 101.57 -9.37000275 30.90800 36.17268
## 123 2020-09-24 3337200 102.44 4.01000214 30.44400 35.70868
## 124 2020-09-25 8336900 113.56 6.07999420 30.15800 35.42268
## 125 2020-09-28 6169800 111.18 -5.72000122 29.91600 35.18068
## 126 2020-09-29 3212000 107.95 0.15999603 30.40600 35.67068
## 127 2020-09-30 4848700 108.35 1.34999847 30.85400 36.11868
## 128 2020-10-01 3704500 104.90 -4.31999969 30.07200 35.33668
## 129 2020-10-02 2731800 103.60 1.90000153 30.47200 35.73668
## 130 2020-10-05 3210200 110.17 5.04000092 30.77800 36.04268
## 131 2020-10-06 2492000 107.41 -1.78999329 31.51567 36.84948
## 132 2020-10-07 1862600 110.00 1.69999695 30.90900 36.24281
## 133 2020-10-08 2352600 111.02 0.06999969 31.17000 36.64519
## 134 2020-10-09 2421500 111.22 -1.08000183 32.59200 38.06719
## 135 2020-10-12 1788200 110.22 -1.02999878 32.12800 37.60319
## 136 2020-10-13 4425900 117.28 5.86999512 31.84200 37.31719
## 137 2020-10-14 2447700 111.76 -7.47000122 31.60000 37.07519
## 138 2020-10-15 2297400 111.42 1.15000153 32.09000 37.56519
## 139 2020-10-16 2118300 107.04 -4.63999939 32.53800 38.01319
## 140 2020-10-19 2708300 102.05 -5.75000000 31.75600 37.23119
## 141 2020-10-20 4039200 98.36 -4.33999634 32.15600 37.63119
## 142 2020-10-21 3398700 91.16 -5.86999512 32.46200 37.93719
## 143 2020-10-22 2571800 93.59 2.98999786 33.19967 38.74891
## 144 2020-10-23 3107000 93.13 0.29000092 32.59300 38.14225
## 145 2020-10-26 2767700 87.38 -3.65000153 32.85400 38.54663
## 146 2020-10-27 2900800 89.59 2.14999390 34.27600 39.96863
## 147 2020-10-28 3414800 81.33 -6.19999695 33.81200 39.50463
## 148 2020-10-29 2815900 87.63 4.68999481 33.52600 39.21863
## 149 2020-10-30 3007300 80.71 -5.52000427 33.28400 38.97663
## 150 2020-11-02 1855200 82.40 0.13000488 33.77400 39.46663
## 151 2020-11-03 1818000 83.64 1.38999939 34.22200 39.91463
## 152 2020-11-04 2272300 87.27 3.16999817 33.44000 39.13263
## 153 2020-11-05 2233100 93.11 3.91000366 33.84000 39.53263
## 154 2020-11-06 1893100 89.86 -2.54000092 34.14600 39.83863
## 155 2020-11-09 6586500 90.31 -4.22000122 34.88367 40.65482
## 156 2020-11-10 6854400 78.74 -6.56000519 34.27700 40.04816
## 157 2020-11-11 3986000 85.01 4.91000366 34.53800 40.45424
## 158 2020-11-12 4953400 90.87 5.10000610 35.96000 41.87624
## 159 2020-11-13 3731700 96.60 3.40999603 35.49600 41.41224
## 160 2020-11-16 6852900 90.65 -5.04999542 35.21000 41.12624
## 161 2020-11-17 2920100 92.89 2.94000244 34.96800 40.88424
## 162 2020-11-18 2525200 89.91 -2.38999939 35.45800 41.37424
## 163 2020-11-19 3249900 86.40 -2.59999847 35.90600 41.82224
## 164 2020-11-20 2557300 86.55 -0.05999756 35.12400 41.04024
## 165 2020-11-23 6548300 93.98 4.99000549 35.52400 41.44024
## 166 2020-11-24 3099200 95.33 0.06999969 35.83000 41.74624
## 167 2020-11-25 3794700 102.60 7.27999878 36.56767 42.56648
## 168 2020-11-27 11777400 125.69 17.59000397 35.96100 41.95982
## 169 2020-11-30 16386900 139.50 20.23999786 36.22200 42.36734
## 170 2020-12-01 12364600 123.47 -26.52999878 37.64400 43.78934
## 171 2020-12-02 3969300 125.48 4.47000122 37.18000 43.32534
## 172 2020-12-03 3234300 129.16 4.18000031 36.89400 43.03934
## 173 2020-12-04 2181800 126.25 -3.11999512 36.65200 42.79734
## 174 2020-12-07 3270700 123.12 -3.77999878 37.14200 43.28734
## 175 2020-12-08 3744200 120.12 -4.47999573 37.59000 43.73534
## 176 2020-12-09 5561800 115.22 -8.40000153 36.80800 42.95334
## 177 2020-12-10 2473300 115.09 4.75999451 37.20800 43.35334
## 178 2020-12-11 6060800 124.88 6.12999725 37.51400 43.65934
## 179 2020-12-14 4690300 129.70 -0.41000366 38.25167 44.48327
## 180 2020-12-15 2727200 126.22 -4.61000061 37.64500 43.87660
## 181 2020-12-16 4096600 120.88 -0.84000397 37.90600 44.28534
## 182 2020-12-17 5070800 131.75 7.26999664 39.32800 45.70734
## 183 2020-12-18 4237400 124.85 -6.12000275 38.86400 45.24334
## lower95 price_change
## 1 8.861959 1.3839998
## 2 10.283959 -0.2080004
## 3 9.819959 2.0059996
## 4 9.533959 2.7820004
## 5 9.291959 1.8239996
## 6 9.781959 3.6140003
## 7 10.229959 3.0359993
## 8 9.447959 4.4079993
## 9 9.847959 4.3680001
## 10 10.153959 3.5920007
## 11 10.905482 3.4243338
## 12 10.298815 5.0109999
## 13 10.469285 9.4699993
## 14 11.891285 5.7780006
## 15 11.427285 6.0820008
## 16 11.141285 5.3779992
## 17 10.899285 5.0699999
## 18 11.389285 5.7600003
## 19 11.837285 4.8519993
## 20 11.055285 3.4939997
## 21 11.455285 2.8139992
## 22 11.761285 1.7080000
## 23 12.501664 1.7003329
## 24 11.894997 2.0270007
## 25 12.056886 1.1259994
## 26 13.478886 0.7139996
## 27 13.014886 1.7579995
## 28 12.728886 7.8140000
## 29 12.486886 23.3759997
## 30 12.976886 23.7359982
## 31 13.424886 20.9179993
## 32 12.642886 27.0300010
## 33 13.042886 39.9599992
## 34 13.348886 36.5940016
## 35 14.078625 34.7663348
## 36 13.471959 32.6329997
## 37 13.626163 28.4120007
## 38 15.048163 29.0499981
## 39 14.584163 26.8140012
## 40 14.298163 26.4700002
## 41 14.056163 27.9120009
## 42 14.546163 34.2019997
## 43 14.994163 30.5739995
## 44 14.212163 25.9659999
## 45 14.612163 25.9759999
## 46 14.918163 27.3099993
## 47 15.637875 25.0623343
## 48 15.031209 26.8090003
## 49 15.178617 25.5779991
## 50 16.600617 23.6260002
## 51 16.136617 25.2299997
## 52 15.850617 31.0159997
## 53 15.608617 32.3380015
## 54 16.098617 35.1079999
## 55 16.546617 38.5200005
## 56 15.764617 44.7819999
## 57 16.164617 45.5819970
## 58 16.470617 48.9860037
## 59 17.180980 56.0883334
## 60 16.574314 62.7350009
## 61 16.715772 56.3239995
## 62 18.137772 61.1220005
## 63 17.673772 61.3259984
## 64 17.387772 57.9619970
## 65 17.145772 60.1439994
## 66 17.635772 57.4540025
## 67 18.083772 82.1259976
## 68 17.301773 76.6480030
## 69 17.701772 74.2480031
## 70 18.007772 72.0020007
## 71 18.709496 81.2243331
## 72 18.102829 87.4810012
## 73 18.239124 88.4000016
## 74 19.661124 96.1180009
## 75 19.197124 116.7820055
## 76 18.911124 114.8079958
## 77 18.669124 117.5100024
## 78 19.159124 122.7799970
## 79 19.607124 115.4719964
## 80 18.825124 110.5939926
## 81 19.225124 115.8640062
## 82 19.531124 124.5580062
## 83 20.224916 120.5803340
## 84 19.618250 122.4469951
## 85 19.750091 118.6660062
## 86 21.172092 130.0139951
## 87 20.708091 131.7779981
## 88 20.422091 148.3840055
## 89 20.180091 142.6360000
## 90 20.670091 144.9359933
## 91 21.118091 152.7079946
## 92 20.336092 124.4599956
## 93 20.736091 98.9999983
## 94 21.042091 107.5539988
## 95 21.728639 120.0463279
## 96 21.121973 129.6729988
## 97 21.249995 120.1119958
## 98 22.671995 119.5199975
## 99 22.207995 116.0540049
## 100 21.921995 110.8299952
## 101 21.679995 92.7120021
## 102 22.169995 86.0720007
## 103 22.617995 87.1340028
## 104 21.835996 80.6459984
## 105 22.235995 80.6359980
## 106 22.541995 82.9299964
## 107 23.221949 76.9323353
## 108 22.615283 75.3590015
## 109 22.740043 75.1780034
## 110 24.162043 63.7060003
## 111 23.698043 56.5499975
## 112 23.412043 62.3659964
## 113 23.170043 64.6280006
## 114 23.660043 66.7279970
## 115 24.108043 76.6300031
## 116 23.326043 81.9519963
## 117 23.726043 91.3419974
## 118 24.032043 86.4160022
## 119 24.706009 78.5283340
## 120 24.099342 80.9250015
## 121 24.221323 82.1439973
## 122 25.643323 70.6619996
## 123 25.179323 71.9960024
## 124 24.893323 83.4019976
## 125 24.651323 81.2640003
## 126 25.141323 77.5439970
## 127 25.589323 77.4959986
## 128 24.807323 74.8280014
## 129 25.207323 73.1279986
## 130 25.513323 79.3919982
## 131 26.181857 75.8943371
## 132 25.575191 79.0910000
## 133 25.694809 79.8499967
## 134 27.116809 78.6280012
## 135 26.652809 78.0920012
## 136 26.366809 85.4379988
## 137 26.124809 80.1600021
## 138 26.614809 79.3299982
## 139 27.062809 74.5020010
## 140 26.280809 70.2940030
## 141 26.680809 66.2040007
## 142 26.986809 58.6980037
## 143 27.650419 60.3903298
## 144 27.043752 60.5369973
## 145 27.161366 54.5259973
## 146 28.583366 55.3139963
## 147 28.119366 47.5180018
## 148 27.833366 54.1039973
## 149 27.591366 47.4259991
## 150 28.081366 48.6260016
## 151 28.529366 49.4179995
## 152 27.747366 53.8299966
## 153 28.147366 59.2700007
## 154 28.453366 55.7140007
## 155 29.112510 55.4263310
## 156 28.505843 44.4629979
## 157 28.621758 50.4720022
## 158 30.043758 54.9100027
## 159 29.579758 61.1039985
## 160 29.293758 55.4400016
## 161 29.051758 57.9219994
## 162 29.541758 54.4520037
## 163 29.989758 50.4940016
## 164 29.207758 51.4260030
## 165 29.607758 58.4560035
## 166 29.913758 59.5000019
## 167 30.568848 66.0323319
## 168 29.962182 89.7290024
## 169 30.076658 103.2780001
## 170 31.498658 85.8260012
## 171 31.034658 88.3000034
## 172 30.748658 92.2660037
## 173 30.506658 89.5980000
## 174 30.996658 85.9780028
## 175 31.444658 82.5300028
## 176 30.662658 78.4120011
## 177 31.062658 77.8819965
## 178 31.368658 87.3659973
## 179 32.020064 91.4483304
## 180 31.413398 88.5750012
## 181 31.526658 82.9739973
## 182 32.948658 92.4220000
## 183 32.484658 85.9859985# the estimated total stock value change between April and August 2020 as result of the pandemic
sum(post_covid_nova$price_change)## [1] 11349.04# Data viz
plot_ly() %>%
add_ribbons(x = post_covid_nova$date,
ymin = post_covid_nova$adj.close_nova,
ymax = post_covid_nova$yhat,
line = list(color = 'rgba(255, 0, 0, 0.05)'),
fillcolor = 'rgba(255, 0, 0, 0.6)',
name = "Estimated Change") %>%
add_segments(x = as.Date("2020-04-01"),
xend = as.Date("2020-04-01"),
y = min(nova$adj.close_nova),
yend = max(nova$adj.close_nova) * 1.05,
line = list(color = "black", dash = "dash"),
showlegend = FALSE) %>%
add_annotations(text = "Pre-Covid19",
x = as.Date("2020-01-01"),
y = max(nova$adj.close_nova) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = "Post-Covid19",
x = as.Date("2020-11-01"),
y = max(nova$adj.close_nova) * 1.05,
showarrow = FALSE) %>%
add_annotations(text = paste("Estimated increase in", "<br>",
"stock price: ~$11,400",
sep = ""),
x = as.Date("2020-08-01"),
y = 3 * 10^1,
arrowhead = 1,
ax = 120,
ay = 40,
showarrow = TRUE) %>%
add_lines(x = nova$date,
y = nova$adj.close_nova,
line = list(color = "#1f77b4"),
name = "Actual") %>%
layout(title = "Covid19 Impact on Stock Value of NOVAVAX",
yaxis = list(title = "Stock Price (US Dollars)"),
xaxis = list(title = "Time Series Model - Auto ARIMA Model",
range = c(as.Date("2019-10-01"), as.Date("2021-01-01"))),
legend = list(x = 0, y = 0.95))