5.2 Candidate model set

Once you have explored a variety of forecasting models you can come up with a candidate set of models along with a set of null models.

Here is our candidate models for the anchovy along with the code to fit and create a forecast from each model.

• Exponential smoothing model with trend

fit <- forecast::ets(traindat, model="AAN")
fr <- forecast::forecast(fit, h=1)

• Exponential smoothing model no trend

fit <- forecast::ets(traindat, model="ANN")
fr <- forecast::forecast(fit, h=1)

• ARIMA(0,1,1) with drift (best)

fit <- forecast::Arima(traindat, order(0,1,1), include.drift=TRUE)
fr <- forecast::forecast(fit, h=1)

• ARIMA(2,1,0) with drift (within 2 AIC of best)

fit <- forecast::Arima(traindat, order(2,1,0), include.drift=TRUE)
fr <- forecast::forecast(fit, h=1)

• Time-varying regression with linear time

traindat$t <- 1:24
fit <- lm(log.metric.tons ~ t, data=traindat)
fr <- forecast::forecast(fit, newdata=data.frame(t=25))

We also need to include null models in our candidate set.

Null models

• Naive no trend

fit <- forecast::Arima(traindat, order(0,1,0))
fr <- forecast::forecast(fit, h=1)
# or simply
fr <- forecast::rwf(traindat, h=1)

• Naive with trend

fit <- forecast::Arima(traindat, order(0,1,0), include.drift=TRUE)
fr <- forecast::forecast(fit, h=1)
# or simply
fr <- forecast::rwf(traindat, drift=TRUE, h=1)

• Average or mean

fit <- forecast::Arima(traindat, order(0,0,0))
fr <- forecast::forecast(fit, h=1)