Exercise 2.1: ts Objects

Year Observation
2012 123
2013 39
2014 78
2015 52
2016 110 
y <- ts(c(123,39,78,52,110), start=2012)
print(y)
## Time Series:
## Start = 2012 
## End = 2016 
## Frequency = 1 
## [1] 123  39  78  52 110
z <- rnorm(24)  # Example random data
y_monthly <- ts(z, start=2003, frequency=12)
print(y_monthly)
##              Jan         Feb         Mar         Apr         May         Jun
## 2003  0.07892140 -0.01494641 -0.60744695 -0.49877288  0.20708524 -0.69320963
## 2004 -1.22666201  1.49370891  0.72677341 -1.64537366  0.21715871 -0.31949408
##              Jul         Aug         Sep         Oct         Nov         Dec
## 2003 -0.27153537  0.74119921 -1.34941574  0.57876384 -1.65858722  1.24228295
## 2004  0.10170494 -0.61251667  0.49055486 -0.98854571 -1.51802061 -0.24212867

Frequency of a Time Series

The “frequency” is the number of observations before the seasonal pattern repeats. Below are common frequencies used in R:

Data Type Frequency
Annual 1
Quarterly 4
Monthly 12
Weekly 52
Daily 7 or 365.25
  • Hourly seasonality (frequency = 60)
  • Daily seasonality (frequency = 1440)
  • Weekly seasonality (frequency = 10080)
  • Annual seasonality (frequency = 525960)
library(fpp3)
install.packages("tsibbledata")
## Warning: package 'tsibbledata' is in use and will not be installed
library(tsibbledata)

autoplot(aus_retail) + ggtitle("Australian Retail Data")
## Plot variable not specified, automatically selected `.vars = Turnover`

Exercise 2.2: Time Plots

# Load required libraries
library(fpp3)
library(tsibbledata)  # Ensure tsibbledata is installed

# Filter for Melbourne-Sydney Economy Class passengers
melsyd <- ansett %>%
  filter(Airports == "MEL-SYD", Class == "Economy")

# Plot the data
autoplot(melsyd, Passengers) + 
  ggtitle("Economy class passengers: Melbourne-Sydney") +
  xlab("Year") +
  ylab("Thousands")

# Load necessary libraries
library(fpp3)
library(tsibbledata)  # Ensure tsibbledata is loaded

# Check available datasets in tsibbledata
data(package = "tsibbledata")

# Visualizing antidiabetic drug sales (PBS dataset)
PBS %>%
  filter(ATC2 == "A10") %>%  # A10 is the code for antidiabetic drugs
  autoplot(Cost) +
  ggtitle("Antidiabetic Drug Sales") +
  ylab("$ million") +
  xlab("Year")

# Exercise 2.3: Time Series Patterns

autoplot(gafa_stock) + ggtitle("Stock Prices of GAFA Companies")
## Plot variable not specified, automatically selected `.vars = Open`

Exercise 2.4: Seasonal Plots

# Load necessary libraries
library(fpp3)

# Use the PBS dataset for antidiabetic drug sales
PBS %>%
  filter(ATC2 == "A10") %>%
  gg_season(Cost, labels = "both") +
  ylab("$ million") +
  ggtitle("Seasonal Plot: Antidiabetic Drug Sales")

# Load necessary libraries
library(fpp3)

# Use the PBS dataset for antidiabetic drug sales
PBS %>%
  filter(ATC2 == "A10") %>%
  gg_season(Cost, polar = TRUE) +
  ylab("$ million") +
  ggtitle("Polar Seasonal Plot: Antidiabetic Drug Sales")
## `geom_line()`: Each group consists of only one observation.
## ℹ Do you need to adjust the group aesthetic?

Exercise 2.5

# Load necessary libraries
library(fpp3)

# Use the PBS dataset for antidiabetic drug sales
PBS %>%
  filter(ATC2 == "A10") %>%
  gg_subseries(Cost) +
  ylab("$ million") +
  ggtitle("Seasonal Subseries Plot: Antidiabetic Drug Sales")

Exercise 2.8

#ggAcf(beer2)

#ggAcf(aelec, lag=48)
library(fpp3)

# Extract beer production data
beer2 <- aus_production %>%
  filter_index("1992 Q1" ~ "2006 Q4") %>%  # Use appropriate years
  select(Beer)

# Compute and plot the ACF for beer production
beer2 %>%
  ACF(Beer) %>%
  autoplot() +
  ggtitle("Autocorrelation Function of Beer Production")