Bootstrap方法在农业产量保险费率厘定中的应用  

The Application of Bootstrap Method in Ratemaking Procedure of Crop Yield Insurance

作　　者：肖宇谷[1] 王克[2] 王晔[3] 

机构地区：[1]中国人民大学统计学院,北京100872 [2]中央财经大学中国精算研究院/中国农业科学院农业信息研究所,北京100871 [3]华农财产保险股份有限公司,北京100037

出　　处：《保险研究》2014年第9期21-28,共8页Insurance Studies

基　　金：教育部人文社会科学重点研究基地项目(13JJD790041)项目成果

摘　　要：农业产量保险的费率厘定常受到数据量少、产量分布选择不一致等问题的影响,使得费率制定者常常对结果的选择感到困惑,缺乏足够的可确信度。为此,本文用Bootstrap方法给出费率的区间估计以弥补点估计的不足。本文通过模拟测试,验证了Bootstrap方法费率点估计和区间估计的优良性。最后,对黑龙江14个县市的玉米产量数据做了实证分析。The premium rating in crop insurance is easily impacted by the uneven yield distribution and the limited data size, and the result often looks uncertain and confusing even to the underwriters themselves. To overcome this shortcoming of point estimation, we used the Bootstrap method to obtain the range estimation for the premium rate. Using a Monte Carlo experiment, the performance of the Bootstrap method for range estimation and point estimation was tested with a favorable effect. Finally, we applied this method to price the corn insurance for 14 counties of Hei- longjiang Province.

关 键 词：农业产量保险 费率厘定 BOOTSTRAP 费率区间估计 

分 类 号：F840.66[经济管理—保险]