Abstract
Tests of goodness of fit of the generalized extreme value (GEV) and the generalized logistic (GL) distribution based on the empirical distribution function are described. Significance points for the Anderson‐Darling (Anderson and Darling, 1952, 1954) and a recently introduced modified Anderson‐ Darling (Sinclair et al., 1987) test statistic are obtained by simulation. For each of these distributions the unknown parameters are estimated from the sample data using probability weighted moments. Approximations for the probability of significance of these test statistics are provided. This enables assessment of GEV and GL flood frequency models for a hydrological region based on all the annual maximum series in the region.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
