Request for a NNETAR() model example with regressors

[fernando-pierucci-alves]

Hello,

NNETAR() is defined and discussed in section 12.4 of the FPP3 book.

Could you please provide an example of this model with regressors?

And if this is doable, it would be helpful if code for plotting the fit against the data were also provided.

Thanks in advance and congrats again for your overall work.

[robjhyndman]

It's the same as any other model with regressors. Just add them on the RHS of the formula. Here is an example.

library(fpp3)
aus <- global_economy |>
    filter(Country == "Australia") 
fit <- aus |>
    model(NNETAR(log(GDP) ~ Population))
fit |>
    forecast(
        new_data = new_data(aus) |> mutate(Population = 24.8e6),
        times = 100
    )
#> # A fable: 1 x 6 [1Y]
#> # Key:     Country, .model [1]
#>   Country   .model                         Year         GDP   .mean Population
#>   <fct>     <chr>                         <dbl>      <dist>   <dbl>      <dbl>
#> 1 Australia NNETAR(log(GDP) ~ Population)  2018 sample[100] 1.38e12   24800000

augment(fit) |>
    ggplot(aes(x = GDP, y = .fitted)) +
    geom_point() +
    scale_y_log10() +
    scale_x_log10() +
    geom_abline(slope = 1, intercept = 0, linetype = "dashed") 

^{Created on 2025-01-05 with reprex v2.1.1}

[fernando-pierucci-alves]

I was able to reproduce it all - thanks!