hp_mix {BayesMortalityPlus}

HP: Model mixture

Description

This function mixes the fitted mortality table of the HP model with another mortality table provided by the user.

Usage

hp_mix (fit, mu_post, weights = NULL, mix_age,
 x0_prior, x0_post, max_age)

Arguments

fit

Object of the class 'HP' fitted by the hp() function.

mu_post

Vector with mortality rates considered in the mix.

weights

Positive vector specifying the weights considered in the mix.

mix_age

Positive vector specifying the age range in the mixture.

x0_prior

Non-negative number indicating the initial age of the fitted HP model.

x0_post

Non-negative number indicating the initial age of the mortality table provided by the user.

max_age

Positive number indicating the final age in the mixture.

Value

Return the posterior distribution for qx.

Examples

## Importing mortality data from the USA available on the Human Mortality Database (HMD):
data(USA)

## Selecting the exposure and death count of the 2010 and 2013 male populations ranging
## from 0 to 90 years old
USA2010 = USA[USA$Year == 2010,]
x = 0:90
Ex = USA2010$Ex.Male[x+1]
Dx = USA2010$Dx.Male[x+1]

USA2013 = USA[USA$Year == 2013,]
Ex2 = USA2013$Ex.Male[x+1]
Dx2 = USA2013$Dx.Male[x+1]

## Fitting HP model for 2010 data and calculating the mortality rates of 2013
fit = hp(x = x, Ex = Ex, Dx = Dx,
         M = 1000, bn = 0, thin = 10)
tx_2013 = 1 - exp(-Dx2/Ex2)

## Mixing fitted model and mortality rates of 2013:
mix <- hp_mix(fit, tx_2013, x0_prior = 0, x0_post = 0, mix_age = c(50,90),
              max_age = 90)

## Obtaining the new estimated mortality table (after mixture):
qx_mix<- apply(mix$qx, 2, median, na.rm = TRUE)
qx_mix