Calculating Insurance Claim Reserves with Hybrid Fuzzy Least Squares Regression Analysis

The prediction of an adequate amount of claim reserves is of the greatest importance to face the responsibilities assumed by an insurance company. Although many different deterministic and stochastic methods based on statistical analyses are used for claims analysis, presence of many internal and external factors that increase the uncertainty in insurance environment may lead to considerable loss in reliability of statistical methods. Therefore, in a state of uncertainty that exist in the nature of many actuarial and financial problems, when convenient and reliable data is not held, the use of fuzzy set theory becomes very attractive to get more actual results. In this paper, a method for calculating insurance claim reserves using hybrid fuzzy least-squares regression analysis is proposed. The results from classical method and this soft computing approach are compared by using original data in automobile liability insurance.