cat(paste("(C) (cc by-sa) Wouter van Atteveldt, file generated", format(Sys.Date(), format="%B %d %Y")))
Note on the data used in this howto: This data can be downloaded from http://piketty.pse.ens.fr/files/capital21c/en/xls/, but the excel format is a bit difficult to parse at it is meant to be human readable, with multiple header rows etc. For that reason, I've extracted csv files for some interesting tables that I've uploaded to https://github.com/vanatteveldt/learningr/tree/master/data. If you're accessing this tutorial from the githup project, these files should be in your 'data' sub folder automatically.
Playing with data in R
To demonstrate R, we will use the data from Piketty's 'Capital in the 21st Century'
income = read.csv("data/income_topdecile.csv")
We've downloaded a csv file and read it into a new variable
income, which should appear in your environment list.
You can click on the file to inspect it visually, but we can also use the
As you can see, the values are NA (missing) for most rows, especially in the earlier period.
Let's throw out all data containing missing values using the
income = na.omit(income) head(income)
Now, we can list the variables in the file using
names and get the numbers of rows or columns with
names(income) nrow(income) ncol(income)
We can also ask for a summary of each of the variables in the file using the
This lists the range, mean, etc. for each variable. We can select any column from a data frame using variable$column:
This gives a vector of numbers representing the different cells in that column.
We can use various functions such as
length to get information about a vector.
length(income$U.S.) mean(income$U.S.) mean(income$Europe)
As perhaps expected, the mean income inequality in Europe is lower than than in the U.S.. Let's do a t-test to see if the difference is significant:
t.test(income$U.S., income$Europe, paired=T)
So, with p<.05 we can conclude that the income distribution in the U.S. is more unequal than in Europe. Let's make a simple plot of the income inequality in the U.S. and Europe (reproducing fig 9.8 on page 324)
plot(x=income$Year, y=income$U.S., type="l", ylab="Top decile income share", xlab="Year", ylim=c(0, 0.5)) lines(x=income$Year, y=income$Europe, col="red")
As you can see, income distribution in pre-WWI Europe is actually more unequal than in the U.S., but this is reversed during the 1910's and inequality diverges after the 1970's. Still, the lines are probably correlated:
So, although the correlation is moderate at 0.43, it is not significant (due to a lack of data points)