model.MEM: The Maximum-Entropy Mortality Model

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

Description

The Maximum-Entropy Mortality Model

Usage

model.MEM(data, x = NULL, y = NULL, n = 5, verbose = FALSE, ...)

Arguments

data	A data.frame or a matrix containing mortality data with ages x as row and time y as column.
x	Numerical vector indicating the ages in input data. Optional. Default: NULL.
y	Numerical vector indicating the years in input data. Optional. Default: NULL.
n	The maximum order of the moments to be used.
verbose	A logical value. Set verbose = FALSE to silent the process that take place inside the function and avoid progress messages.
...	Arguments to be passed to or from other methods.

Details

\insertNoCite

pascariu_phd2018MortalityForecast

Value

The output is an object of class MEM with the components:

input	List with arguments provided in input. Saved for convenience;
info	Short details about the model;
call	An unevaluated function call, that is, an unevaluated expression which consists of the named function applied to the given arguments;
coefficients	Estimated coefficients;
fitted.values	Fitted values of the estimated model;
observed.values	Observed values used in fitting;
fitted.raw.moments	Fitted raw moments;
observed.raw.moments	Observed raw moments of the input data;
residuals	Deviance residuals;
random.walk.model	Object containing the components of the fitted time series model to the extrapolate moments;
x	Vector of ages used in the fitting;
y	Vector of years used in the fitting.

References

\insertAllCited

See Also

predict.MEM plot.MEM find.moments

Examples

# Data
x  <- 0:110
y  <- 1985:2014
dx <- HMD_male$dx$GBRTENW[paste(x), paste(y)]

# Fit model the Maximum-Entropy Mortality of order 5,
# that is using the first 6 statistical moments (moment 0 to 5).

M <- model.MEM(data = dx, x = x, y = y, n = 5)
M
R <- residuals(M)

plot(M, plotType = "observed")
plot(M, plotType = "fitted")

plot(R, plotType = "scatter")
plot(R, plotType = "colourmap")
plot(R, plotType = "signplot")

# Perform forecasts
P <- predict(M, h = 16, x.h = 0:120)

plot(P, plotType = "mean")
plot(P, plotType = "lower")
plot(P, plotType = "upper")

plot(P, M, plotType = "raw_moments")
plot(P, M, plotType = "normalised_moments")

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