Produces forecasts from a trained model.

# S3 method for TSLM
forecast(
  object,
  new_data,
  specials = NULL,
  bootstrap = FALSE,
  approx_normal = TRUE,
  times = 5000,
  ...
)

Arguments

object

A model for which forecasts are required.

new_data

A tsibble containing the time points and exogenous regressors to produce forecasts for.

specials

(passed by fabletools::forecast.mdl_df()).

bootstrap

If TRUE, then forecast distributions are computed using simulation with resampled errors.

approx_normal

Should the resulting forecast distributions be approximated as a Normal distribution instead of a Student's T distribution. Returning Normal distributions (the default) is a useful approximation to make it easier for using TSLM models in model combinations or reconciliation processes.

times

The number of sample paths to use in estimating the forecast distribution when bootstrap = TRUE.

...

Other arguments passed to methods

Value

A list of forecasts.

Examples

as_tsibble(USAccDeaths) %>%
  model(lm = TSLM(log(value) ~ trend() + season())) %>%
  forecast()
#> # A fable: 24 x 4 [1M]
#> # Key:     .model [1]
#>    .model    index            value .mean
#>    <chr>     <mth>           <dist> <dbl>
#>  1 lm     1979 Jan t(N(8.9, 0.003)) 7620.
#>  2 lm     1979 Feb t(N(8.8, 0.003)) 6899.
#>  3 lm     1979 Mar t(N(8.9, 0.003)) 7639.
#>  4 lm     1979 Apr   t(N(9, 0.003)) 7841.
#>  5 lm     1979 May t(N(9.1, 0.003)) 8645.
#>  6 lm     1979 Jun t(N(9.1, 0.003)) 9087.
#>  7 lm     1979 Jul t(N(9.2, 0.003)) 9908.
#>  8 lm     1979 Aug t(N(9.1, 0.003)) 9237.
#>  9 lm     1979 Sep   t(N(9, 0.003)) 8237.
#> 10 lm     1979 Oct   t(N(9, 0.003)) 8516.
#> # ℹ 14 more rows