Change Variable for Autoplot for Simple Forecast

[JohnSparks2024]

Hi,
I altered the example code for figure 5.7 to forecast the Electricity variable. Unfortunately, for the autoplot it plots the actual values of the Beer variable in the forecast time period and gives me the note: 'Plot variable not specified, automatically selected .vars = Beer'. I presume it selects this because it is the leftmost variable in the ts_tibble. Unfortunately, I don't see how to alter the variable selected for this graph. Would you please let me know how to do so? My code is shown below. Thanks very much. --John Sparks

train <- aus_production |>
  filter_index("1992 Q1" ~ "2006 Q4")
# Fit the models
Elec_fit <- train |>
  model(
    Mean = MEAN(Electricity),
    `Naïve` = NAIVE(Electricity),
    `Seasonal naïve` = SNAIVE(Electricity),
    Drift = NAIVE(Electricity ~ drift())
  )
ElecFitMatrix<-augment(Elec_fit)

# Generate forecasts for 14 quarters
Elec_fc <- Elec_fit |> forecast(h = 14)

# Plot forecasts against actual values
Elec_fc |>
  autoplot(train, level = NULL) +
  autolayer(
    filter_index(aus_production, "2007 Q1" ~ .),
    colour = "black"
  ) +
  labs(
    y = "Electricity",
    title = "Forecasts for quarterly Electricity production"
  ) +
  guides(colour = guide_legend(title = "Forecast"))

[robjhyndman]

You need to add .vars = Electricity as an argument in autolayer(). If you don't specify a variable, it chooses the first variable that is not either an index or a key variable.