predict.LiLee: Forecast age-specific death rates using the Li-Lee model

In mpascariu/MortalityForecast: Standard Tools to Compare and Evaluate Various Mortality Forecasting Methods

Description

Forecast age-specific death rates using the Li-Lee model

Usage

## S3 method for class 'LiLee'
predict(object, h, order.B = c(0, 1, 0),
  include.drift.B = TRUE, order.D = c(1, 0, 0),
  include.drift.D = FALSE, level = c(80, 95),
  jumpchoice = c("actual", "fit"), method = "ML", verbose = TRUE,
  ...)

Arguments

object	An object of class LiLee.
h	Number of years to be forecast in the future.
order.B	The ARIMA order for the benchmark population model. This is A specification of the non-seasonal part of the ARIMA model: the three components (p, d, q) are the AR order, the degree of differencing, and the MA order. If order.B = NULL, the ARIMA order will be estimated automatically using the KPPS algorithm;
include.drift.B	Logical. Should we include a linear drift term in the ARIMA of benchmark population model?
order.D	The ARIMA order driving the deviation from the benchmark population model. If order = NULL, this will be estimated automatically.
include.drift.D	Logical. Should we include a linear drift term in the ARIMA driving the deviation from the benchmark?
level	Significance level of the confidence interval.
jumpchoice	Method used for computation of jumpchoice. Possibilities: "actual" (use actual rates from final year) and "fit" (use fitted rates).
method	ARIMA fitting method: maximum likelihood or minimize conditional sum-of-squares. Options to use: conditional-sum-of-squares ("CSS-ML"), maximum likelihood ("ML") and "CSS".
verbose	A logical value. Set verbose = FALSE to silent the process that take place inside the function and avoid progress messages.
...	Additional arguments to be passed to Arima

Details

\insertNoCite

li2005MortalityForecast

Value

The output is a list with the components:

call	An unevaluated function call, that is, an unevaluated expression which consists of the named function applied to the given arguments;
predicted.values	A list containing the predicted values together with the associated prediction intervals given by the estimated model over the forecast horizon h;
conf.intervals	Confidence intervals for the extrapolated kt parameters;
kt.arima	An object of class ARIMA that contains all the components of the fitted time series model used in kt prediction;
kt	The extrapolated kt parameters;
x	Vector of ages used in prediction;
y	Vector of years used in prediction;
info	Short details about the model;
benchmark	An object containing the predicted results for the benchmark population.

Author(s)

Marius D. Pascariu and Marie-Pier Bergeron-Boucher

References

\insertAllCited

See Also

model.LiLee

Examples

# For examples go to ?model.LiLee

mpascariu/MortalityForecast documentation built on Sept. 28, 2020, 2:40 p.m.