A Comparative Study of Rainfall Estimates Using Six Statistical Distributions

摘要

Statistical analysis of historical rainfall, involving fitting of an appropriate distribution for obtaining rainfall intensities of different interval, is very much useful for making resource allocation decisions concerning risks. This paper gives details of a study involving the use of six statistical distributions for rainfall estimation for Barh and Megharaj sites. Anderson Darling test statistic was used for checking the goodness of fit of the distributions to the recorded data. Diagnostic analysis, involving D-index statistic, was used for evaluating the suitability of the probability distribution for the data under study. The study shows the Log Pearson Type III to be the best suited, among six statistical distributions studied, for extreme value analysis of the rainfall for the data under study.