偏正态与偏t分布对巨灾损失数据的拟合分析

The Fitting Analysis of Skew-Normal and Skew-t Distribution to the Catastrophe Loss Data

如何更好地拟合巨灾损失数据一直是一个重要的问题。本文将偏正态分布和偏t分布应用到实际巨灾损失数据中,选取了福州市近8年火灾造成的经济损失为样本数据,采用极大似然估计的方法对参数进行估计,并且选取常用来拟合损失的分布进行比较。结果表明,偏t分布最适合拟合巨灾损失数据。How to better fit catastrophe loss data is always an important problem. This paper applies the skew-normal distribution and skew-t distribution to the actual catastrophe loss data, and selects the economic loss caused by fire in Fuzhou in recent 8 years as the sample data. Besides, this paper uses the maximum likelihood estimation method to estimate the parameters, and selects the commonly used loss distribution for comparison. The results show that skew t distribution is most suitable for fitting catastrophe loss data.