学会誌「自然災害科学」

シミュレーションデータに基づく波高の極値統計解析モデルの精度の検討

Investigation on Accuracy of Extreme Wave Analysis Methods Using Numerically Simulated Data

山口 正隆*・大木 泰憲**・中村 雄二***

Masataka YAMAGUCHI*, Yasunori OHKI** and Yuji NAKAMURA***

Abstract

In the statistical analysis of extreme wave height data, many attempts have ever been made to find what kind of fitting method is preferable for the parameter estimation of a probability distribution to obtain a reliable estimate of return wave height and how the sampling variability could be evaluated, but the answer is still uncertain, because the class of parent distribution and the parameter condition investigated are limited.

This study uses 8 kinds of probability distributions including the Weibull and Gumbel distributions and 4 kinds of parameter estimation methods such as the probability weighted moment method and maximum likelihood method. Based on the statistical analysis with use of 3 or 4 methods for the data generated by a Monte-Carlo simulation technique, the optimum method to be used for parameter estimation was determined for each distribution from the view points of minimizing bias and variance of the return wave heights. Also applicabilities of the jackknife method and information matrix method to the estimation of variance of return wave height were verified.

キーワード:

Key words: