robjhyndman

library(fpp3)
state_tourism <- tourism |>
  group_by(State) |>
  summarise(Trips = sum(Trips))
training <- state_tourism |>
  filter(year(Quarter) <= 2014)
fit <- training |>
  model(
    ets = ETS(Trips),
    arima = ARIMA(Trips),

robjhyndman

library(fpp3)

# Stretched data
stretch_uschange <- us_change |>
  stretch_tsibble(.init = 5)

# Fit models to all folds of data
fit <- stretch_uschange |>
  model(lm = TSLM(Consumption ~ Income))

robjhyndman

par(mar=c(0,0,0,0))
plot(0,0,xlim=c(0,28),ylim=c(0,1),
       xaxt="n",yaxt="n",bty="n",xlab="",ylab="",type="n")
i <- 1
for(j in 1:20)
{
  test <- (6+j):26
  train <- 1:(5+j)
  arrows(0,1-j/20,27,1-j/20,0.05)
  points(train,rep(1-j/20,length(train)),pch=19,col="blue")

robjhyndman

library(forecast)

fc <- function(y, h, xreg, newxreg) {
  fit <- auto.arima(y, xreg=xreg)
  forecast(fit, xreg=newxreg, h=h)
}

y <- ts(rnorm(100))
x <- matrix(ts(rnorm(100)),ncol=1)
tsCV(y, fc, xreg=x, h=1)

robjhyndman

par(mar=c(0,0,0,0))
plot(0,0,xlim=c(0,28),ylim=c(0,1),
       xaxt="n",yaxt="n",bty="n",xlab="",ylab="",type="n")
i <- 1
for(j in 1:20)
{
  test <- (6+j):26
  train <- 1:(5+j)
  arrows(0,1-j/20,27,1-j/20,0.05)
  points(train,rep(1-j/20,length(train)),pch=19,col="blue")

robjhyndman

library(forecast)
fit <- auto.arima(USAccDeaths, xreg=fourier(USAccDeaths, 4))
fc <- forecast(fit, xreg=fourier(USAccDeaths, 4, 12))
plot(fc)
for(i in 1:12)
  lines(simulate(fit, xreg=fourier(USAccDeaths, 4, 12)))