3.1 Transformations and adjustments

To simplify the patterns in the historical data by removing known sources of variation some adjustments are required:

  • calendar adjustments
  • population adjustments (data per person (or per thousand people, or per million people) rather than the total)
  • inflation adjustments (such as price index adjustments)
  • mathematical transformations (log/power/Box-Cox transformation, …)

Features by package Guerrero’s method for Box Cox lambda selection

library(fpp3)


lambda <- aus_production %>%
  features(Gas, features = guerrero) %>%
  pull(lambda_guerrero)


aus_production %>%
  autoplot(box_cox(Gas, lambda)) +
  labs(y = "",
       title = latex2exp::TeX(paste0(
         "Transformed gas production with $\\lambda$ = ",
         round(lambda,2))))+
  ggthemes::theme_pander()