Family expenditure data, heteroscedasticity and the Law of Demand

This paper presents some results of a data analysis of Family Expenditure Surveys for the United Kingdom and France. It is shown that these data sets exhibit characteristic features that we call “increasing spread” and “increasing dispersion”. Statistical tools used for analysing the data are discussed. They are based on concepts from non-parametric regression and average derivative estimation. One can show that increasing dispersion supports the weak axiom of revealed preferences of the market demand function. Increasing spread provides empirical evidence for the Law of Demand.