Prognoser i R
Rob J. Hyndman
Professor of Statistics at Monash University
set.seed(3) # Reproducibility
wn <- ts(rnorm(36)) # White noise
autoplot(wn) # Plot!

"Vitt brus" är en tidsserie av iid-data
ggAcf(wn) +
ggtitle("Sample ACF for white noise")

ggAcf(wn) +
ggtitle("Sample ACF for white noise")

ggAcf(wn) +
ggtitle("Sample ACF for white noise")

ggAcf(wn) +
ggtitle("Sample ACF for white noise")

pigs <- window(pigs, start=1990)
autoplot(pigs/1000) +
xlab("Year") +
ylab("thousands") +
ggtitle("Monthly number of pigs slaughtered in Victoria")

ggAcf(pigs) +
ggtitle("ACF of monthly pigs slaughtered
in Victoria")

ggAcf(pigs) +
ggtitle("ACF of monthly pigs slaughtered
in Victoria")

ggAcf(pigs) +
ggtitle("ACF of monthly pigs slaughtered
in Victoria")

Ljung-Box-testet tar hänsyn till de första h autokorrelationsvärdena tillsammans.
Ett signifikant test (litet p-värde) tyder på att data troligen inte är vitt brus.
Box.test(pigs, lag = 24, fitdf = 0, type = "Lj")
Box-Ljung test
data: pigs
X-squared = 634.15, df = 24, p-value < 2.2e-16
Prognoser i R