广义极值分布参数估计方法的对比分析

简介了广义极值分布的3种参数估计方法:极大似然(ML)、线性矩(LM)和间隔最大积(MPS)方法的特点和计算方法,采用历年马口月最大径流量和广州日最大降水量作为广义极值分布不同参数估计方法的实证分析例子。分析结果表明,两实例各自3种参数估计方法得到的3个参数值较为接近,各种拟合优度检验结果表明两实例均服从广义极值分布,但MPS参数估计推算的设计值与观测值拟合更好。

Comparative Analysis of Parameter Estimation Methods of Generalized Extreme Value Distribution

Three parameter estimation methods of generalized extreme value distribution function were briefly introduced in the paper, which included the maximum likelihood estimation, the linear moment estimation and the maximum product of spacing estimation. Two demonstration examples including monthly maximum runoff at Makou Station and daily maximum precipitation in Guangzhou over the past years were analyzed by the three parameter estimation methods of GEV distribution. The results indicated that three parameters obtained by three different estimation methods were very close. Several goodness fit tests showed that two examples were obeyed the generalized extreme value distribution. And the designed values predicted by maximum product of spacing estimation were better fitted with the measured values.