Applies a fitted NNETAR model to a new dataset.

# S3 method for NNETAR
refit(object, new_data, specials = NULL, reestimate = FALSE, ...)

Arguments

object The time series model used to produce the forecasts A tsibble containing future information used to forecast. (passed by fabletools::forecast.mdl_df()). If TRUE, the networks will be initialized with random starting weights to suit the new data. If FALSE, for every network the best individual set of weights found in the pre-estimation process is used as the starting weight vector. Additional arguments for forecast model methods.

Value

A refitted model.

Examples

lung_deaths_male <- as_tsibble(mdeaths)
lung_deaths_female <- as_tsibble(fdeaths)

fit <- lung_deaths_male %>%
model(NNETAR(value))

report(fit)
#> Series: value
#> Model: NNAR(3,1,2)[12]
#>
#> Average of 20 networks, each of which is
#> a 4-2-1 network with 13 weights
#> options were - linear output units
#>
#> sigma^2 estimated as 21523
fit %>%
refit(new_data = lung_deaths_female, reestimate = FALSE) %>%
report()
#> Series: value
#> Model: NNAR(3,1,2)[12]
#>
#> Average of 20 networks, each of which is
#> a 4-2-1 network with 13 weights
#> options were - linear output units
#>
#> sigma^2 estimated as 236049