Use X-12-Arima And Sarima Model And Combined Model Predicting Chinese Premium Income

Along with the swift development of economy and the rapid growth of the national income, there is a growing inhabitants’ demand for insurance, resulting in the increasing trend of the premium income also year by year. Therefore, it is very necessary to find the scientific method to accurately predict the premium income. Due to the obvious seasonal characteristic of the premium income series, this article employs the first seasonal index method and X-12method to capture this character of premium income, the results show that X-12method shows superiority over the first seasonal index method in analysis of the seasonal characteristics. To better predict the growth of premium income, this paper firstly establishes the seasonal difference auto-regressive moving average (SARIMA) model, and the X-12multiplication model--ased auto-regressive moving average model (X-12-multiplication ARIMA model) and X-12additive model-based auto-regressive moving average model (X-12-ARIMA model). Through the comparison between these models, it is found that SARIMA model and X-12-ARIMA multiplication model is superior to X-12-ARIMA additive model. Then, this paper combines SARIMA multiplication model and X-12-ARIMA model to the final prediction of the premium income. In this combinational model, the particle swarm optimization algorithm is used to optimize the weight of two models. Finally we make empirical analysis to the monthly premium income of the major insurance companies of China between January1999and June1999, and predict the trend of the premium income in our country, thereby providing necessary support for the insurance industry and the regulation to insurance industry.