Estimators based on sample quantiles using (h,phi)-entropy measures

Abstract

A point estimation procedure based on the maximum entropy principle for (h, phi) entropies is proposed using sample quantiles. These estimators are efficient and asymptotically normal under standard regularity conditions. A test for goodness-of-fit is constructed, being the corresponding statistic asymptotically distributed chi-squared. These results generalize the results obtained in [1].