The Research Of Loss Reserve Evaluation For Non-life Insurance

The debt management is the biggest difference between insurance industry and other industries. Insurance companies must serve sufficient policy reserves to meet its obligations under the compensation for policies that have been sold. It will get a high attention from insurance companies and authorities that whether the policy reserve is reasonable and adequately estimated. Of all the policy reserves, the Reserve for Outstanding Losses is the hardest to estimate. The withdrawal Reserve for Outstanding Losses is collected to make up for the future insurance accident that may have happened. Because it is to be done in the future and affected by many uncertain factors, it’s even more difficult to be estimated.Most of the outstanding claims for non-life insurance reserves are based on its assessment of theoretical research methods, there have not been much discussion about their risk estimation. Outstanding claim reserve risk assessment mainly refers to its volatility, resulting in outstanding claims reserve volatility by many factors and mainly divided into two kinds:internal and external. Internal risks include risk data, risk assessment, Claims policy risks and risk management; external risks include the risk of inflation and other macroeconomic factors. Some of these risks will not directly affect the outstanding claims reserves estimates impact, but indirectly by influencing.This paper first makes a detailed description and analysis about the existing assessment methods and assessment risk of the outstanding loss reserve on the basis of the research results now available both at home and abroad, and then makes an empirical analysis on the risk that is quantifiable by the stochastic GLM method, and based on the results of these analysis, puts forward the proposals to improve the accuracy and to prevent volatility of the reserve evaluation.This paper is divided into five chapters:The first chapter is the introduction part. Firstly, it aims to get the importance of the reserve evaluation from an introduction to the importance of responsibilities provisioning; secondly, makes a brief analysis on the current assessment methods and assessment risk of the non-life business reserve; then analyses the present research status and development trend of the reserve assessment methods; and at last points out its deficiencies.The second chapter introduces the assessment methods of non-life insurance claim reserves in detail. The assessment method is divided into two categories, the general method and the stochastic method. The general method, which is prescribed by the CIRC, includes the case estimate method, the premium ratio method, the case law compensation method, the reserve progressive method and the B-F method. Whereas, Mark model, linear regression model, lognormal model and generalized linear model are briefly introduced when the stochastic method is mentioned. And the two types of assessment methods are evaluated in detail, their deficiencies are also pointed out in this chapter.The third chapter describes non-life claim reserves assessment risk in detail. The internal and external risks of the non-life claim reserves will be carried out in a detailed description, and the reasons that caused these risks will also be analyzed.The fourth chapter is the theoretical basis of the empirical analysis part. This chapter first briefly introduces the assessment GLM, including the Poisson model and the Gamma model, and their prediction error estimation of the non-life claims reserve. And then the article aims to get the double generalized linear model (DGLM), where the Claim Data are included in the calculation, and its prediction error estimation through the combination of the GLM theory and the compensation method.The fifth chapter is the empirical analysis part. This chapter, by using the data from Swiss Mobile Insurance, mainly analysis the risk of wasting data information and economic inflation, which are quantifiable by using GLM and DGLM methods. On the one hand, this chapter, by making a comparison on the prediction error obtained by using GLM and DGLM methods, gets the conclusion that the prediction error is significantly smaller by using DGLM method, and this means that when the Claim Data are included in the calculation, non-life insurance reserve estimates are of better applicability. On the other hand, this chapter, by making a comparison of the prediction error of the outstanding loss reserve estimates obtained before and after the treatment of inflation, gets the conclusion that the outstanding loss reserve estimate is significantly smaller after the treatment inflation, and this proves that after the treatment of inflation, the outstanding reserve estimates are of better applicability.The sixth chapter is the conclusion part. In this chapter, the proposals to prevent the risk of volatility of non-life insurance outstanding reserves are put forward based on the analysis on risk assessment and the empirical analysis in the previous chapters.The innovation of this paper not only lies in the detailed introduction of research methods and assessment risks for non-life insurance reserves, but also in the empirical analysis for the quantitative risks of all assessment risks by using GLM methods. And at last, based on the results of the above research, this article gives out suggestions on the assessment to non-life insurance reserves in China.