normal.test: Normality check procedure for residuals
In eik4862/Subway: Time Series Analysis & Visualization

Description Usage Arguments Value Author(s) Examples

Take fitted ARIMA model and perform normality checks. It runs three well-known normality test, Shapiro-Wilk test, Jarque-Bera test, and Kolmogorov-Smirnov test and prints out the result to the console. If plot is TRUE, it also plots Q-Q plot of residuals and estimated density of residual for visual check.

1	normal.test(fit, alpha = 0.05, plot = TRUE)

`fit`	Fitted ARIMA model(`ARIMA` object).
`alpha`	The probability of type 1 error for normality tests. (default = 0.05)
`plot`	If `TRUE`, the Q-Q plot of residuals and estimated density of residuals will be presented. Otherwise, not. (default = TRUE)

A list containing the following elements:

`shapiro.wilk`	The test result of Shapiro-Wilk test(`htest` objects).

`jarque.bera`	The test result of Jarque-Bera test(`htest` objects).

`komogorov.smirnov`	The test result of Kolmogorov-Smirnov test(`htest` objects).

`plot`	`ggplot` object of the plot. (Returned only if `plot` is TRUE.)

Sanghyun Park, Daun Jeong, and Sehun Kim

# 228 is the station code for SNU
data <- get.subway(228)

# Fit SARIMA model with total on and off passenger counts
SNU.total <- ts(data$total, frequency = 7)
SNU.fit <- auto.arima(SNU.total)

# Normality check
normal.test(SNU.fit)