Life Insurance Demand Influencing Factors Analyze By Ridge Regression And Forecast BP Neural Network

Insurance industry has been developed rapidly since China restarted its insurance service in 1979.As a business related to people's livelihood,life insurance is an important part of the insurance industry.Since life insurance was resumed in 1982,it has made a big progress.The income from Chinese life insurance was only 5.08 billion in 1990,and yet it reached 2223.46 billion yuan in 2016.Compared with the income in 1990,that in 2016 has increased by more than 300 times.Since 2000 in particular,with the emergence of various investments in life insurance products,more and more consumers begin to buy these products.Life insurance has gradually become the main source of the income in insurance industry.Therefore,analyze the factors that stimulate the demand of life insurance service so that life insurance premium income can be reasonably calculated in.For relevant departments,such analysis can also provide certain reference to make correlate development planning.Life insurance demand can be measured by the income of life insurance.According to the relevant data of 21 year from 1996 to 2016 as well as the related study of the scholars at home and abroad,the qualitative study can be applied into the thesis to analyze the factors in economy,population and policy environment firstly and use the regression analysis method to conduct the quantitative analysis.The results show that life insurance demand is positively correlated with GDP per capita,savings,income,education and urbanization level,while bank rate is negatively correlated with the demand of life assurance,the relation between the rate of mortality and inflation as well as aging and life insurance demand and the inflation rate is not obvious.Then,these factors can be used as a basis of the input index of the neural network to establish the BP neural network forecast model of life insurance premium income.The comparison between the series of error evaluation index of ridge regression and the result of BP neural network prediction indicates that the effect of BP neural network model of prediction is superior to the ridge regression model.Based on the comparison above,the BP neural network prediction model can be much better.The thesis can be employed to predict life assurance income in 2017 and 2018.