HW#1
Hunter Patenaude
08-31-21
Load packages and data
library(fpp3)
library(ggplot2)
library(tidyverse)
library(dplyr)
library(tsibble)
library(moments)
library(tibbletime)
library(lubridate)
# install and load any package necessaryQuestions
Exercise 1
tute1 <- readxl::read_excel("Downloads/tute1.xlsx",
col_types = c("date", "numeric", "numeric",
"numeric"))
mytimeseries <- tute1 %>%
mutate(Quarter = yearmonth(Quarter)) %>%
as_tsibble(index = Quarter)
mytimeseries %>%
pivot_longer(-Quarter) %>%
ggplot(aes(x = Quarter, y = value, colour = name)) +
geom_line()
mytimeseries %>%
pivot_longer(-Quarter) %>%
ggplot(aes(x = Quarter, y = value, colour = name)) +
geom_line()+
facet_grid(name ~ ., scales = "free_y")
Exercise 2
Exercise 3
tourismvs <- readxl::read_excel("//Users//hunterpatenaude//Downloads//tourism.xlsx")
tourismmm <- tourismvs %>%
mutate(Quarter = yearquarter(Quarter)) %>%
as_tsibble(key = c("Region", "State", "Purpose"),index = "Quarter")
tt<-tourismvs %>% group_by(Region,Purpose)%>%
summarise (mean = mean(Trips)) %>% ungroup() %>% filter (mean == max(mean))## `summarise()` has grouped output by 'Region'. You can override using the
## `.groups` argument.as.data.frame(tourismmm) %>% group_by(Region,Purpose) %>%
summarise (mean = mean(Trips)) %>% ungroup() %>% filter (mean == max(mean))## `summarise()` has grouped output by 'Region'. You can override using the
## `.groups` argument.## # A tibble: 1 × 3
## Region Purpose mean
## <chr> <chr> <dbl>
## 1 Sydney Visiting 747.tourismvs %>% group_by(State) %>% summarise (total = sum(Trips))## # A tibble: 8 × 2
## State total
## <chr> <dbl>
## 1 ACT 41007.
## 2 New South Wales 557367.
## 3 Northern Territory 28614.
## 4 Queensland 386643.
## 5 South Australia 118151.
## 6 Tasmania 54137.
## 7 Victoria 390463.
## 8 Western Australia 147820.Exercise 4
ausarr <- aus_arrivals
autoplot(ausarr)## Plot variable not specified, automatically selected `.vars = Arrivals`
gg_season(ausarr)## Plot variable not specified, automatically selected `y = Arrivals`
gg_subseries(ausarr)## Plot variable not specified, automatically selected `y = Arrivals`
The main observation that I noticed was that arrivals in the UK always fell during Q2 and Q3. During the 10 year span between 2000 Q1 and 2010 Q1 New Zealand had a huge spike of arrivals and Japan had a huge drop in arrivals.
Exercise 5
set.seed(6547)
ausser <- aus_retail %>%
filter(`Series ID` == sample(aus_retail$`Series ID`,1))
autoplot(ausser)## Plot variable not specified, automatically selected `.vars = Turnover`
gg_season(ausser)## Plot variable not specified, automatically selected `y = Turnover`
gg_subseries(ausser)## Plot variable not specified, automatically selected `y = Turnover`
gg_lag(ausser)## Plot variable not specified, automatically selected `y = Turnover`
ausser %>%
ACF(Turnover) %>%
autoplot()
Using this seed I can see a clear trend in turnovers. As time moves forward, there is a positive trend of turnover. Each yeah the amount of turnovers seem to increase. Seasonality seems to exists, because the amount of turnovers tends to spike around every December. I don’t see any evidence of the data being cyclic.
Exercise 6
library(tsibbledata)
fff <- as.data.frame(gafa_stock) %>% group_by(Symbol) %>%
filter(Symbol == "FB")
vec1 <- pull(fff,Close)
FBmean <- mean(vec1, na.rm = TRUE)
FBmedian <- median(vec1)
FBsd <- sd(vec1, na.rm = TRUE)
FBkurt <- kurtosis(vec1, na.rm = TRUE)
FBskew <- skewness(vec1, na.rm = TRUE)
vecdiff <- diff(vec1)
FBmeandif <- mean(vecdiff, na.rm = TRUE)
FBsddif <- sd(vecdiff, na.rm = TRUE)
FBkurtdif <- kurtosis(vecdiff, na.rm = TRUE)
FBskewdif <- skewness(vecdiff, na.rm = TRUE)
FBmeanbh <- sum(vec1, na.rm = TRUE) / nrow(fff)
FBsdbh <- sqrt(sum((vec1-mean(vec1))^2/(length(vec1)-1)))
FBkurtbh <- ((sum((vec1-mean(vec1))^4))/length(vec1))/(sum((vec1-mean(vec1))^2)/length(vec1))^2
FBskewbh <- (3*(FBmeanbh-FBmedian))/FBsdbhFBmean## [1] 120.4625FBmeanbh## [1] 120.4625FBmeandif## [1] 0.06076372FBsd## [1] 41.32364FBsdbh## [1] 41.32364FBsddif## [1] 2.414555FBkurtbh## [1] 1.836712FBkurtdif## [1] 74.02921FBskewbh## [1] 0.2023634FBskewbh## [1] 0.2023634FBskewdif## [1] -3.973192Excercise 7
TSLA_stock <- readxl::read_excel("Downloads/TSLA stock.xlsx",
col_types = c("date", "numeric", "numeric",
"numeric", "numeric", "numeric",
"numeric"))
tesla <- TSLA_stock %>% select(-(Open:Close)) %>%
select(-(Volume))
teslats <- tesla %>%
mutate(Date = as_date(Date)) %>%
as_tsibble(index = "Date", key = "Adj Close")
teslaline <- teslats %>%
filter(Date >= as.Date("2022-06-01") & Date <= as.Date("2022-06-30")) %>%
ggplot(aes(x = Date, y = `Adj Close`)) +
geom_line()
teslaline
teslaJAN <- teslats %>%
filter(Date >= as.Date("2022-01-01") & Date <= as.Date("2022-01-31"))
teslaFEB <- teslats %>%
filter(Date >= as.Date("2022-02-01") & Date <= as.Date("2022-02-28"))
teslaMAR <- teslats %>%
filter(Date >= as.Date("2022-03-01") & Date <= as.Date("2022-03-31"))
teslaAPR <- teslats %>%
filter(Date >= as.Date("2022-04-01") & Date <= as.Date("2022-04-30"))
teslaMAY <- teslats %>%
filter(Date >= as.Date("2022-05-01") & Date <= as.Date("2022-05-31"))
teslaJUN <- teslats %>%
filter(Date >= as.Date("2022-06-01") & Date <= as.Date("2022-06-30"))
teslaJUL <- teslats %>%
filter(Date >= as.Date("2022-07-01") & Date <= as.Date("2022-07-31"))
teslaAUG <- teslats %>%
filter(Date >= as.Date("2022-08-01") & Date <= as.Date("2022-08-31"))
TJAN_Vec <- pull(teslaJAN, "Adj Close")
TJAN_M <- mean(TJAN_Vec)
TFEB_Vec <- pull(teslaFEB, "Adj Close")
TFEB_M <- mean(TFEB_Vec)
TMAR_Vec <- pull(teslaFEB, "Adj Close")
TMAR_M <- mean(TMAR_Vec)
TAPR_Vec <- pull(teslaAPR, "Adj Close")
TAPR_M <- mean(TAPR_Vec)
TMAY_Vec <- pull(teslaMAY, "Adj Close")
TMAY_M <- mean(TMAY_Vec)
TJUN_Vec <- pull(teslaJUN, "Adj Close")
TJUN_M <- mean(TJUN_Vec)
TJUL_Vec <- pull(teslaJUL, "Adj Close")
TJUL_M <- mean(TJUL_Vec)
TAUG_Vec <- pull(teslaAUG, "Adj Close")
TAUG_M <- mean(TAUG_Vec)
TJAN_var <- var(TJAN_Vec)
TFEB_var <- var(TFEB_Vec)
TMAR_var <- var(TMAR_Vec)
TAPR_var <- var(TAPR_Vec)
TMAY_var <- var(TMAY_Vec)
TJUN_var <- var(TJUN_Vec)
TJUL_var <- var(TJUL_Vec)
TAUG_var <- var(TAUG_Vec)#JAN - SD and Mean
TJAN_var; TJAN_M## [1] 1003.339## [1] 336.7228#FEB - SD and Mean
TFEB_var; TFEB_M## [1] 270.8465## [1] 292.9616#MAR - SD and Mean
TMAR_var; TMAR_M## [1] 270.8465## [1] 292.9616#APR - SD and Mean
TAPR_var; TAPR_M## [1] 622.5522## [1] 332.4625#MAY - SD and Mean
TMAY_var; TMAY_M## [1] 891.3576## [1] 255.2233#JUN - SD and Mean
TJUN_var; TJUN_M## [1] 126.5048## [1] 234.0259#JUL - SD and Mean
TJUL_var; TJUL_M## [1] 387.2411## [1] 251.3947#AUG - SD and Mean
TAUG_var; TAUG_M## [1] 91.80111## [1] 294.8699