ilc: extract.deaths – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

extract.deaths

Miscellaneous utility functions for demogdata type mortality data

Description

This function calculates and outputs the corresponding (observed) number of deaths from a demogdata type mortality data for a choice of ages and calendar years.

Usage

extract.deaths(data, ages = data$age, years = data$year, combine.upper = T, 
				fill.method = NULL, series = names(data$rate)[1])

Arguments

`data`	mortality data object of `demogdata` class
`ages`	vector of ages to extract
`years`	vector of years to extract
`combine.upper`	logical, if TRUE, ages above max(ages) will be grouped together
`fill.method`	string value indicating the method to be used for correcting missing or 0 transition rates before estimating the number of deaths (e.g. one of "perks", "interpolate" or "mspline"). By default is set to NULL, which corresponds to no data correction.
`series`	target series name (e.g. 'males') or index number (e.g. 1) of the data object to be extracted

Value

A 'fictive' demogdata class object in which the (mortality) rate component is replaced by the extracted number of deaths.

Note

When estimating the number of deaths (as the product between mortality rates and exposures), some assumptions will need to be made for the cases where the mortality rates are missing (NA). That is, it is not possible to estimate the number of deaths where the exposure (population) is zero because for those cases the corresponding hazard rate (mu) will most likely be NA in the dataset (unless mu was estimated by other means, like a moving average or smoothing, etc.). However, it is reasonable to assume that zero exposures correspond to no observed deaths, which is implemented here. Further, when a fill.method is specified, then the zero and the missing mortality rates are corrected before calculating the number of deaths.

Author(s)

Z. Butt and S. Haberman and H. L. Shang

Examples

# 'observed' number of deaths (i.e. no data correction)
extract.deaths(dd.cmi.pens, ages=55:100)
# number of deaths with corrections using Perks mortality model
tmp.d <- extract.deaths(dd.cmi.pens, ages=55:100, fill='perks')
# Note: to further improve the plot the user can change the vertical
# axis label and/or main title (amongst other plotting parameters).
plot(tmp.d, transf=FALSE, ylab='Number of Deaths')    # change ylab
plot_dd(tmp.d, transf=FALSE, ylab='Number of Deaths', lpar=list(x.int=-0.2, y.int=0.9, cex=0.85))
plot_dd(tmp.d, y=1995:2003, transf=FALSE, lty=1:5, ylab='Number of Deaths',
  main=paste(tmp.d$lab, ": male (1995-2003)", sep='')) # change main title

ilc

Lee-Carter Mortality Models using Iterative Fitting Algorithms

v1.0

GPL (>= 2)

Authors

Zoltan Butt, Steven Haberman and Han Lin Shang

Initial release

2014-11-19