These systems are used in cross-selling industries, and they measure correlated items as well as their user rate. This last point wasn’t included the apriori algorithm (or association rules), used in market basket analysis.

I explore the sampling distributions of estimates of Gini coefficients from a sample, using the New Zealand Income Survey 2011. At the actual sample size of nearly 30,000, sampling error is negligible, and as low as a sample size of 1,000 a 95% confidence interval is (0.48, 0.55), precise enough for most purposes and certainly good enough given the non-sampling vagaries of the underlying data. I discuss why individual and weekly income data – which is all I have to hand – returns a higher measure of inequality than does annual household income, the more usual and internationally-comparable (and completely valid) measure.