HW1

Load packages and data

# install and load any package necessary
library(fpp3)
library(moments)
library(tidyverse)
library(tidyquant)
library(USgas)
tute1 <- readr::read_csv("C:\\Users\\PythonAcct\\Downloads\\tute1.csv")
tourism1 <- readxl::read_excel("C:\\Users\\PythonAcct\\Downloads\\tourism.xlsx")
gme <- readr::read_csv("C:\\Users\\PythonAcct\\Downloads\\GME (3).csv")
# install and load any package necessary

Questions

Exercise 1

tute2=tute1 %>%
  mutate(Quarter=yearquarter(Quarter)) %>%
  as_tsibble(index=Quarter)
ggplot(tute2, aes(Quarter, Sales, group=1,color='blue')) + geom_line()

ggplot(tute2, aes(Quarter, AdBudget, group=1,color='blue')) + geom_line()

ggplot(tute2, aes(Quarter, GDP, group=1,color='blue')) + geom_line()

tute3 <- tute2 %>% 
  pivot_longer(cols=c("Sales","AdBudget","GDP"),
               names_to = "Measure",
               values_to="Dollars")
ggplot(data=tute3) + 
  geom_line(mapping = aes(x = Quarter, y = Dollars, color=Measure)) + 
  facet_grid(~ Measure)

Exercise 2

totals <- us_total %>% 
  as_tsibble(index=year,key=state)
totals %>% 
  filter(state=="New Hampshire"|state=="Maine"|state=="Vermont"|state=="Massachusetts"|state=="Connecticut"|state=="Rhode Island") %>%
  autoplot(y/1e3)+
  labs(y="Natural Gas Consumption (Thousands)", x="Time")

Exercise 3

The output of true signifies that the constructed tsibble and given tsibble are the same

Adelaide, Business is the highest average combination.

tourism2 <- tourism
tourism3 <- tourism1 %>%
  mutate(Quarter=yearquarter(Quarter)) %>%
  as_tsibble(index=Quarter,key=c(Region,State,Purpose))
all.equal(tourism3,tourism2)

## [1] TRUE

tourism4 <- tourism1 %>%
  mutate(Quarter=yearquarter(Quarter)) %>%
  as_tsibble(index=Quarter,key=c(Region,State,Purpose)) %>%
  group_by(Region,Purpose) %>%
  summarise(Avg=mean(Trips)) %>% 
  ungroup() %>%
  mutate(Max=max(Avg))
tourism4

## # A tsibble: 24,320 x 5 [1Q]
## # Key:       Region, Purpose [304]
##    Region   Purpose  Quarter   Avg   Max
##    <chr>    <chr>      <qtr> <dbl> <dbl>
##  1 Adelaide Business 1998 Q1  135.  985.
##  2 Adelaide Business 1998 Q2  110.  985.
##  3 Adelaide Business 1998 Q3  166.  985.
##  4 Adelaide Business 1998 Q4  127.  985.
##  5 Adelaide Business 1999 Q1  137.  985.
##  6 Adelaide Business 1999 Q2  200.  985.
##  7 Adelaide Business 1999 Q3  169.  985.
##  8 Adelaide Business 1999 Q4  134.  985.
##  9 Adelaide Business 2000 Q1  154.  985.
## 10 Adelaide Business 2000 Q2  169.  985.
## # … with 24,310 more rows
## # ℹ Use `print(n = ...)` to see more rows

tourism5 <- tourism1 %>%
  mutate(Quarter=yearquarter(Quarter)) %>%
  as_tsibble(index=Quarter,key=c(Region,State,Purpose)) %>%
  group_by(State) %>%
  summarise(Sum=sum(Trips)) %>% 
  ungroup()
tourism5

## # A tsibble: 640 x 3 [1Q]
## # Key:       State [8]
##    State Quarter   Sum
##    <chr>   <qtr> <dbl>
##  1 ACT   1998 Q1  551.
##  2 ACT   1998 Q2  416.
##  3 ACT   1998 Q3  436.
##  4 ACT   1998 Q4  450.
##  5 ACT   1999 Q1  379.
##  6 ACT   1999 Q2  558.
##  7 ACT   1999 Q3  449.
##  8 ACT   1999 Q4  595.
##  9 ACT   2000 Q1  600.
## 10 ACT   2000 Q2  557.
## # … with 630 more rows
## # ℹ Use `print(n = ...)` to see more rows

Exercise 4

aus_arrivals %>%
  autoplot(Arrivals)

The only irregularity I see is the divergence after around 2001. Japan also plummets after 1995.

aus_arrivals %>%
  gg_season(Arrivals)

I don’t really see anything here.

aus_arrivals %>%
  gg_subseries(Arrivals)

This graph makes Japan’s data look quite strange. It is the only one that exhibits the stark concavity.

Exercise 5

set.seed(1888274)
myseries <- aus_retail %>%
  filter(`Series ID` == sample(aus_retail$`Series ID`,1))
myseries %>%
  autoplot()

## Plot variable not specified, automatically selected `.vars = Turnover`

myseries %>%
  gg_season()

## Plot variable not specified, automatically selected `y = Turnover`

myseries %>% 
  gg_subseries

## Plot variable not specified, automatically selected `y = Turnover`

myseries %>%
  gg_lag()

## Plot variable not specified, automatically selected `y = Turnover`

myseries %>% 
  ACF(Turnover) %>% 
  autoplot

In the first plot, there is a large spike around 2008.

In the second plot, there is always a large spike at the end of the year. The late 2000s are the highest.

The third plot proves the increased numbers for December.

The fourth plot provides a lot of insight.

The fifth plot exhibits a downward trend.

Exercise 6

Part 1

mystock <- gafa_stock %>% 
  filter(Symbol=="FB") %>% 
  pull(Close)
mean(mystock)

## [1] 120.4625

sd(mystock)

## [1] 41.32364

Part 2

len <- seq(1:nrow(gafa_stock %>% filter(Symbol=="FB")))
len2 <- seq(1:(nrow(gafa_stock %>% filter(Symbol=="FB"))-1))
first_dif = c()
for (i in len2){
  first_dif[i]=mystock[i+1]-mystock[i]
}
mean(first_dif)

## [1] 0.06076372

sd(first_dif)

## [1] 2.414555

skewness(first_dif)

## [1] -3.973192

kurtosis(first_dif)

## [1] 71.02921

Part 3

fb_avg <- (sum(mystock))/nrow(gafa_stock %>% filter(Symbol=="FB"))
len <- seq(1:nrow(gafa_stock %>% filter(Symbol=="FB")))
total_var=0.0
for (i in len){
  total_var=total_var+(((mystock[i]-fb_avg)^2)/length(len))
}
fb_sd <- sqrt(total_var)
total_skew=0.0
for (i in len){
  total_skew=total_skew+(((mystock[i]-fb_avg)^3)/length(len))
}
fb_skew=total_skew/((sd(mystock))^3)
total_kurt=0.0
for (i in len){
  total_kurt=total_kurt+(((mystock[i]-fb_avg)^4)/length(len))
}
fb_kurt=total_kurt/((fb_sd)^4)
fb_avg

## [1] 120.4625

fb_sd

## [1] 41.30721

fb_skew

## [1] 0.2276621

fb_kurt

## [1] 1.836712

Exercise 7

gme <- readr::read_csv("C:\\Users\\PythonAcct\\Downloads\\GME (3).csv")

## Rows: 169 Columns: 7
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## dbl  (6): Open, High, Low, Close, Adj Close, Volume
## date (1): Date
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

g1 <- gme %>%
  mutate(n=row_number()) %>% 
  mutate(Day=as.Date(Date)) %>%
  as_tsibble(index=n) %>% 
  select("n","Day","Adj Close")
g2 <- g1 %>% 
  filter(month(Day)==6)
autoplot(g2)

## Plot variable not specified, automatically selected `.vars = Adj Close`

g22 <- g1 %>% 
  as_tibble()
g3 <- g22 %>% 
  group_by(month(Day)) %>%
  summarise(avg=mean(`Adj Close`),var=var(`Adj Close`))
g3

## # A tibble: 9 × 3
##   `month(Day)`   avg     var
##          <dbl> <dbl>   <dbl>
## 1            1  29.5 19.6   
## 2            2  29.2  5.45  
## 3            3  29.9 67.4   
## 4            4  36.3  9.09  
## 5            5  26.6 16.0   
## 6            6  32.7  3.52  
## 7            7  34.2  7.12  
## 8            8  36.6 18.1   
## 9            9  27.5  0.0364

HW1

Brandon Pascali

2022-09-06

Load packages and data

Questions

Exercise 1

Exercise 2

Exercise 3

The output of true signifies that the constructed tsibble and given tsibble are the same

Adelaide, Business is the highest average combination.

Exercise 4

The only irregularity I see is the divergence after around 2001. Japan also plummets after 1995.

I don’t really see anything here.

This graph makes Japan’s data look quite strange. It is the only one that exhibits the stark concavity.

Exercise 5

In the first plot, there is a large spike around 2008.

In the second plot, there is always a large spike at the end of the year. The late 2000s are the highest.

The third plot proves the increased numbers for December.

The fourth plot provides a lot of insight.

The fifth plot exhibits a downward trend.

Exercise 6

Part 1

Part 2

Part 3

Exercise 7