Birth and death rates in China 1949-2010

This document shows the net change in the population by subtracting death rate from birth rate over the period 1949-2010. In particular, note the negative effect on the size of the population during the time of the 'Great Leap Forward' in the late 1950s. The change in population appears now to be stabilising.

Load the data which contains variables for Year, Population, Birth and Death Rates and Births Less Death.

china <- read.csv("http://dl.dropbox.com/u/23281950/china.csv")

We'll need the 'zoo' package for working with time-series

install.packages("zoo")

## Installing package(s) into 'C:/Users/Stephen/Documents/R/win-library/2.15'
## (as 'lib' is unspecified)

## Error: trying to use CRAN without setting a mirror

library(zoo)

## Attaching package: 'zoo'

## The following object(s) are masked from 'package:base':
## 
## as.Date, as.Date.numeric

Make two plots side by side. One is of Births Less Deaths, and the other is of Population. The 'BLD' plot has a horizontal line at zero, which is where more people are dying than are being born. In other words, the population is shrinking.First, make objects containing the plot data

cpop <- zoo(china$Population, china$Year)

cbld <- zoo(china$BLD, china$Year)

Now run the two plots. One is of the birth less deaths and the second is population over time. We can get them side by side with

par(mfrow = c(2, 1))
plot(cbld, ylab = "Births Minus Deaths", main = "China: births less deaths", 
    lwd = 2, col = "blue")
grid()
abline(a = 0, b = 0, lwd = 3, col = "red")
plot(cpop, ylab = "Population", main = "China: population", lwd = 2, col = "red")
grid()

It seems reasonable to think that there must be some correlation between BLD and Population. We can find this using cross-correlation

ccf(china$Population, china$BLD, main = "Births less Deaths")

Statistically significant lags between the two time series are those below the blue line. Until about ten years there is statistically significant correlation. Try the same but using only the birth rate

ccf(china$Population, china$Birth.Rate, main = "Birth Rate ")

There is an interesting symmetric lag….