Forecast demogdata data using Lee-Carter method.

The kt coefficients are forecast using a random walk with drift. The forecast coefficients are then multiplied by bx to obtain a forecast demographic rate curve.

Usage

# S3 method for lca
forecast(
  object,
  h = 50,
  se = c("innovdrift", "innovonly"),
  jumpchoice = c("fit", "actual"),
  level = 80,
  ...
)

Arguments

object: Output from lca.
h: Number of years ahead to forecast.
se: Method used for computation of standard error. Possibilities: “innovdrift” (innovations and drift) and “innovonly” (innovations only).
jumpchoice: Method used for computation of jumpchoice. Possibilities: “actual” (use actual rates from final year) and “fit” (use fitted rates).
level: Confidence level for prediction intervals.
...: Other arguments.

Value

Object of class fmforecast with the following components:

label: Region from which the data are taken.
age: Ages from object.
year: Years from object.
rate: List of matrices containing forecasts, lower bound and upper bound of prediction intervals. Point forecast matrix takes the same name as the series that has been forecast.
fitted: Matrix of one-step forecasts for historical data

Other components included are

e0: Forecasts of life expectancies (including lower and upper bounds)
kt.f: Forecasts of coefficients from the model.
type: Data type.
model: Details about the fitted model

Author

Rob J Hyndman

Examples

france.lca <- lca(fr.mort, adjust="e0")
france.fcast <- forecast(france.lca, 50)
plot(france.fcast)

plot(france.fcast,'c')