This paper compares various parameter estimation methods for three hydrologic frequency analysis models: three-parameter log-normal (3LN), the Gumbel (EV1), and the generalized extreme-value (GEV) distributions. Ten methods for 3LN including the maximum likelihood, moments, probability weighted moments (PWM), Ishihara-Takase, improved Iwai, sextiles, and least-squares are evaluated through Monte Carlo simulation studies in terms of the root mean square error (RMSE) of the estimates of the quantiles (T-year events). Six methods for EV1 and four methods for GEV are compared in the same manner. The merits of the methods are discussed in relation to the sample size N.
