ABSTRACT

In this chapter, the authors explains two or several groups of observations rather than a single set, and analyzes to compare the distributions of the groups. They consider the gross national products of all countries in the United Nations in 1980, but separated into northern hemisphere and southern hemisphere countries. The authors describe the empirical quantile- quantile plot. The shows how certain kinds of derived plots based on differences and ratios can enhance our ability to perceive structure in the data. The decline is most dramatic for the highest quantiles. A sensitive method for comparing two distributions is the empirical quantile-quantile plot, which allows a detailed comparison over the entire range of the distributions. Distributions can differ in such interesting and varied ways that one or two summary statistics are often simply inadequate. Several data distributions can be compared by making a collection of empirical quantile-quantile plots, one for each pair of data sets.