Bayesian estimation of the Bonferroni index from a Pareto-type I population
The Bonferroni index (B) is a measure of income and wealth inequality, and it is particularly suitable for poverty studies. Since most income surveys are of a sample nature, we propose Bayes estimators ofB from a Pareto/I population. The Bayesian estimators are obtained assuming a squared error loss function and, as prior distributions, the truncated Erlang density and the translated exponential one. Two different procedures are developed for a censored sample and for income data grouped in classes.
Key wordsBonferroni inequality index Bayes estimator Pareto/I distribution truncated Erlang distribution translated exponential distribution squared error loss function
Unable to display preview. Download preview PDF.