摘要
本文从《中国保险年鉴》和中国保监局广东省监管局网站获取广东省(不含深圳)财产保险业统计数据,应用移动平均值法和适应性预期法对赔付率进行平滑;鉴于赔付率数据能够较好拟合差分自回归移动平均模型,采用小波降噪去除白噪声的方法,平滑赔付率数据。三种方法中,小波降噪平滑后的数据与原数据误差最小,平滑效果最好。三种平滑方法均能够抵消巨额未决赔款拉低总产出的影响,较好地平滑了财产保险业的总产出。
Using the data excerpted from China Insurance Yearbook and the website of Authority of Guangdong Province （excluding Shenzhen District） , China Insurance Regulatory Commission, moving average approach and a- doptive expectation approach are adopted to smooth the claim rate of non-life insurance in Guangdong Province. For the time series of claim rate fits with ARIMA model quite well, wavelet denoising method is also used to smooth claim rate. Wavelet denoising approach receives better outcome than other two approaches with the smallest smoot- hing error from the real claim rate. All these methods could efficiently count the sudden change of pending payment and reduce the fluctuation in the output of non-life insurance in Guangdong Province.
出处
《华南理工大学学报（社会科学版）》
2013年第6期10-14,共5页
Journal of South China University of Technology(Social Science Edition)
基金
广东省哲学社会科学规划项目(GD10XYJ11)
中央高校基本科研业务费专项基金(2013XZD06)
广东省普通高校人文社科重点研究基地项目(x2jmN910019a)
教育部人文社会科学研究一般项目(11YJC790038)
