Discussion Papers in Statistics and Econometrics No. 5/96 [rev.]
University of Cologne, Seminar of Economic and Social Statistics, Cologne
Most empirical studies on poverty consider the extent of poverty either for the entire society or for separate groups like elderly people.However, these papers do not show what the situation looks like for persons of a certain age. In this paper poverty measures depending on age are derived using the joint density of income and age. The density is nonparametrically estimated by weighted Gaussian kernel density estimation. Applying the conditional density of income to several poverty measures provides an insight into the extent of poverty over the life cycle. Furthermore confidence bands for the poverty age profiles were computed.