Journal of Applied and Natural Science 1(1):50-52(2009)
Best fitting probability distributions for annual maximum discharge data of the river Kopili, Assam
Abhijit Bhuyan and Munindra Borah*
Department of Mathematical Sciences, Tezpur University, Napaam, Tezpur-784 028, INDIA
*Corresponding author. E-mail : email@example.com
Abstract : In this study our main objective is to determine the best fitting probability distribution for annual maximum flood discharge data of river Kopili, Assam. Various probability distributions i.e. Gumbel (G), generalized extreme value (GEV), normal (N), log-normal (LN3), generalized logistic (GLO), generalized pareto (GPA) and Pearson type-III (PE3) have been used for our study. The L-moments methods have been used for estimating the parameters of all the distributions. The root mean square error (RMSE), model efficiency and D-index (fit in the top six values) together with L-moment ratio diagram is used for goodness of fit measure. It has been observed that Generalized Pareto is the best fitting probability distribution for annual maximum discharge data of river Kopili.
Keywords : L-moments, Probability distributions, Quantiles, L-moment ratio diagram