plot_chain {BayesMortalityPlus}

Chain's plot

Description

This function provides three options of plots for the chain generated by the MCMC algorithm in hp() and dlm() functions.

Usage

plot_chain(fit, param, type = c("trace", "acf", "density"))

Arguments

fit

Object of the classes HP or DLM.

param

Character vector specifying the parameters to be plotted. It is used only when the class of fit object is DLM.

type

Character string specifying the type of plot to be returned. There are three options: "trace" return a plot for the sample of the parameters; "acf" return a plot for the autocorrelation of the parameters; "density" return a plot for the posterior density of the parameters based on the samples generated by the MCMC method.

Value

A plot of the chosen type of the selected parameter(s).

Examples

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

## Selecting the log mortality rate of the 2010 total population ranging from 0 to 90 years old
USA2010 = USA[USA$Year == 2010,]
x = 0:90
Ex = USA2010$Ex.Total[x+1]
Dx = USA2010$Dx.Total[x+1]
y = log(Dx/Ex)

## Fitting HP model
fit = hp(x = x, Ex = Ex, Dx = Dx, model = "lognormal",
         m = c(NA, 0.08, rep(NA, 6)),
         v = c(NA, 1e-4, rep(NA, 6)))

## Plotting all the available types of plot:
plot_chain(fit, type = "trace")
plot_chain(fit, type = "acf")
plot_chain(fit, type = "density")


## Fitting DLM
fit = dlm(y, M = 100)

plot_chain(fit, param = "sigma2", type = "trace")
plot_chain(fit, param = "mu[10]", type = "acf")

## Selecting all theta1 indexed with 1 in first digit
plot_chain(fit, param = "theta1[1", type = "density")

## Plotting all parameters indexed by age 10 and age 11
plot_chain(fit, param = c("[10]", "[11]"))