Skip to contents

Fit a functional principal components model

Source: R/HU.R
fpc_model.Rd

Model mortality or fertility data using functional principal components as proposed by Hyndman & Ullah (2007) Missing rates are set to the geometric mean rate for the relevant age.

Usage

fpc_model(.data, rates, order = 6)

Arguments

.data

A vital object including an age variable and a variable containing mortality or fertility rates.

rates

Variable in `.data` containing mortality or fertility rates. If omitted, it will search for a variable with one of the following names: `mx`, `mortality`, `fx`, `fertility` or `rate` (not case sensitive).

order

Number of principal components to fit.

Value

A list of class fpc_model containing various model objects.

References

Hyndman, R.J., and Ullah, S. (2007) Robust forecasting of mortality and fertility rates: a functional data approach. Computational Statistics & Data Analysis, 51, 4942-4956. https://robjhyndman.com/publications/funcfor/

Author

Rob J Hyndman

Examples

# Compute Hyndman-Ullah model for Australian females, males and total
aus_hu <- aus_mortality |>
  dplyr::filter(Code == "AUS") |>
  fpc_model()
aus_hu
#> Functional data model
#> 
#> Sub-groups:
#> # A tibble: 3 × 4
#>   Sex    State     Code  varprop
#>   <chr>  <chr>     <chr>   <dbl>
#> 1 female Australia AUS     0.980
#> 2 male   Australia AUS     0.970
#> 3 total  Australia AUS     0.984
autoplot(aus_hu) +
  patchwork::plot_annotation("Hyndman-Ullah components for Australia")