基于变结构协整理论的保费预测模型  

A Premium Forecasting Model Based on Variable Structure Cointegration Theory

作　　者：陆秋君[1,2] 陈玲[1] 施锡铨[2] 

机构地区：[1]上海理工大学理学院,上海200093 [2]上海财经大学金融学院,上海200433

出　　处：《数学的实践与认识》2011年第3期1-7,共7页Mathematics in Practice and Theory

摘　　要：提出了一种基于变结构协整理论的保费预测建模新方法,所建模型反映了保费和GDP之间的长期静态和短期动态波动的均衡关系.通过确定时间序列突变点,并利用突变点信息提高模型的预测精度,避免了传统的保费预测中经常存在的虚假回归问题.采用该方法对中国年度保费进行了预测分析,结果表明了该方法的有效性.The author proposes a new method for premium forecasting modeling basea on variable structure cointegration theory.The established model reflects the equilibrium relation of long-term static state and short-term dynamic fluctuation between premium and GDP.By determining the points of discontinuity in time series and using the information of discontinuity points,the forecasting accuracy of the model is improved.Meanwhile the pseudoregression that often presents in traditional premium forecasting can be avoided.Using the method,the forecasting and analysis of annual premium in China are carried out,and the results show that the method is available.

关 键 词：参数变结构协整 虚假回归 保险市场 

分 类 号：O242.1[理学—计算数学]