A nonparametric approach to k-sample inference based on entropy

Entropy as a measure of uncertainty is no longer restricted to the domain of communication theory. It is being used in several branches of statistics. In this paper we consider nonparametric methods of estimation of entropy. Using nonparametric methods, we also develop a test of the hypothesis of equality of entropy for multiple groups. A simulation study is performed to compare the power of the proposed test with existing parametric and nonparametric procedures. Finally a bootstrap distribution of the proposed test statistic is considered for two data sets as illustrative examples.