带双误差Lee-Carter模型在新农保中的运用

发布时间：2018-05-07 12:24

本文选题：带双误差Lee-Carter模型 + 逐年递推模型　；参考：《杭州电子科技大学》2014年硕士论文

【摘要】：国家财政补助是新农保和老农保最大的区别，财政补贴能否持续到位，是新农保体系能否持续发展的关键。所以研究新农保的收支情况，对分析国家财政支持力度意义重大。农村人口的预期寿命与新农保的收支模型有着密切的关系。随着人口死亡率的下降，这意味着人口的预期寿命在不断延长。国际流行趋势说明采用动态的死亡率模型研究人口的预期寿命，能更加精确地预测出未来人口寿命。 Lee-Carter模型是经典的动态死亡率模型，本文首先对该模型进行改进，将原模型中的单误差模型改为双误差模型，分别用来度量不同年龄、不同时间所引起的死亡率波动，从而在一定程度上解决经典Lee-Carter模型死亡率预测偏高问题。然后本文分析我国农村男女死亡率现状以及发展趋势，并利用带双误差的Lee-Carter模型来预测未来农村人口分性别、分年龄的死亡率，进而预测出到2020年各年人口结构及人口预期寿命。最后本文分析新农保的养老金集资模式与收支情况，建立逐年递推的宏观收入支出模型，利用模型分析在不同档次的养老保险、不同补交概率下国家财政支出情况，并根据预测的财政支出情况，提出相关建议。本文分为五章。第一章阐述本文研究背景、研究意义、死亡率模型和新农保收支模型的研究现状。第二章对生命表、死亡率预测模型、经典Lee-Carter模型进行简单阐述，针对原Lee-Carter模型的误差项进行详细分析，并在原模型的基础上提出了改进的模型—带双误差的Lee-Carter模型，把误差项分解成两个独立的随机误差项，使得改进后的Lee-Carter模型的方差比原模型小，且这两个误差项是可以估计的。第三章分析国家财政在新农保中的责任所在，同时，，介绍参保者的缴费原则、领取养老金原则。本章从宏观角度出发，利用递推方式构建新农保养老金的收入模型和养老金支出模型，用于预算未来我国财政在新农保中支出情况。第四章利用中国农村历年人口死亡率数据来拟合带双误差的Lee-Carter模型，并用于预测未来人口死亡率以及人口预期寿命，进而利用新农保逐年递推收支模型预算未来国家财政在新农保中的收支情况。第五章为本文的结论。
[Abstract]:The state financial subsidy is the biggest difference between the new rural insurance and the old one. Whether the financial subsidy can be sustained is the key to the sustainable development of the new rural insurance system. Therefore, studying the income and expenditure of new rural insurance is of great significance to the analysis of the state financial support. The life expectancy of rural population is closely related to the income and expenditure model of new rural insurance. As the death rate declines, this means that life expectancy is increasing. The international epidemic trend shows that using dynamic mortality model to study population life expectancy can more accurately predict future population life expectancy. The Lee-Carter model is a classical dynamic mortality model. In this paper, we first improve the model by changing the single error model into a double error model, which is used to measure the mortality fluctuation caused by different age and time. In order to solve the problem of high mortality prediction of classical Lee-Carter model to some extent. Then this paper analyzes the present situation and development trend of male and female mortality in rural areas, and uses the Lee-Carter model with double errors to predict the future rural population mortality by sex and age. Furthermore, the population structure and life expectancy of each year by 2020 are predicted. Finally, this paper analyzes the new rural pension funding model and income and expenditure situation, establishes a recursive macro-income and expenditure model year by year, uses the model to analyze the state financial expenditure under different levels of pension insurance and different replenishment probability. And according to the forecast financial expenditure situation, puts forward the related proposal. This paper is divided into five chapters. The first chapter describes the research background, significance, mortality model and new rural insurance income and expenditure model. In the second chapter, the life table, mortality prediction model and classical Lee-Carter model are briefly described. The error terms of the original Lee-Carter model are analyzed in detail. Based on the original model, an improved Lee-Carter model with double errors is proposed. The error term is decomposed into two independent random error terms, which makes the variance of the improved Lee-Carter model smaller than that of the original model, and these two error terms can be estimated. The third chapter analyzes the responsibility of the state finance in the new rural insurance, at the same time, introduces the contributory principle and the pension principle of the insured. From the macro point of view, this chapter constructs the income model and the pension expenditure model of the new rural insurance pension by using the recursive method, which is used to budget the future expenditure of our country's finance in the new rural insurance. The fourth chapter uses the Chinese rural population mortality data to fit the Lee-Carter model with double errors, and uses it to predict the future population mortality and population life expectancy. Then, the budget of the future state finance in the new rural insurance is estimated by using the recursive income and expenditure model of the new rural insurance year by year. The fifth chapter is the conclusion of this paper.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F842.67;F323.89;C921

【参考文献】