| In order to ensure that the insurance company has enough funds to pay for outstanding claims reserves and ensure the financial stability of the company in the future,it is necessary to keep an appropriate reserves from the income in the accident year.This paper presents Bayes chain ladder model which is based on payments per claim incurred.The ideas from credibility theory are used and credibility estimates of chain ladder factors are derived under generalized weighted loss function.Fur-thermore,we get estimates of reserves.Finally,practical examples of insurance company are given and show that credibility claims reserving which is based on the generalized weighted loss function is valid.Considering that if there is a large claim,in order to make the prediction result,more accurat,e,we can use the balanced loss function to penalize the difference.For making sure estimate of accident average reserve is not dependent on the concrete form of prior distribution,B-F reserve model adopts the method of credibility the-ory.On the basis of considering the claim amounts in the same accident progress in different years have a certain relationship,paper give accident claims the inhomoge-neous and homogeneous credibility estimates under balanced loss function.Under the same assumption of the original model and under balanced loss function,we get,estimates of reserves.In particular,when ω= 0,the result is same as the original model.Due to the flow of the triangle data sample size is limited,the quantile regression model can directly calculate the outstanding claims reserve and its sensitivity to outliers in the data is not high,can get a more accurate prediction results.Using regression model to estimate reserves,it can solve the limited problem of claim data sample size.On the basis of the Meng Sheng’s model,the quantile regression model combined with t,he credibility theory,we get the estimates of reserves and give the parameter estimation. |
