README.md

Lifetable Data in R

Stephen Wade

May 2016

Introduction

Provides long-form lifetable datasets in R which are currently sourced from:

| Source | R name | Copyright | |---------------------------------|----------|----------------------------------| | Australian Bureau of Statistics | abs_lt | © Commonwealth of Australia |

Use of any of these data sources should be attributed/cited as per the instructions provided by the original source given in the preceding table.

The data is in long form which is identified by sex, age, location, year and the quantity type e.g. lx or qx and so on. In R the data looks like:

library(lifetable)
head(abs_lt)
  sex measure age     value loc year
1   M      lx   0 100000.00  AU 2002
2   M      lx   1  99467.41  AU 2002
3   M      lx   2  99420.44  AU 2002
4   M      lx   3  99392.67  AU 2002
5   M      lx   4  99369.61  AU 2002
6   M      lx   5  99351.11  AU 2002

Installation

From R (using github installation via devtools):

library(devtools)
install_github('stephematician/lifetable')

Usage

Example Queensland 2010-2012

To select the Queensland 2010-2012 lifetable from the ABS dataset, abs_lt:

qld_2010 <- select_lifetable(abs_lt, 2010, 'QLD')
head(qld_2010)
  measure sex age  value
1      lx   F   0 100000
2      lx   F   1  99554
3      lx   F   2  99526
4      lx   F   3  99503
5      lx   F   4  99487
6      lx   F   5  99475

Data

Australian Bureau of Statistics

The ABS website provides explanatory notes for all their Australian lifetable data (link).

One important constraint is that the pre-2009 data is not broken bown by state and territory, whereas 2010 onwards data provides a national lifetable as well as individual state and territory lifetables.

Source

The lifetable data was downloaded from the following ABS web-pages accessed on 12-13th May 2016 for data up to 2012--2014:

Cleaning/manual work

The final files in the data-raw/abs directory were manually modified for the purpose of loading and creating the final ABS dataset: abs_lt.

File naming

Some raw files were renamed to match the later file-name convention:

File formats

The 2001-2003 and 2003-2005 files were in an old unsupported Excel format (version 5.0) and so was 'saved as' in Excel 97-2003 Workbook (.xls) format.

Lx column

Some values for the 100 and over Lx measurement were given as (a)9654 in order to refer to a footnote. The (a) is not part of the number and was manually edited out so that the number could be read into R.

Empty rows

The Excel file for 2004-2006 was large due to a high number of empty rows, these were removed via the delete cells option.



stephematician/lifetable documentation built on May 30, 2019, 3:17 p.m.