The essential feature of an insurance practice is to set the premium at the beginning of the insurance contract. To determine the correct premium for next year in an insurance company, precise and reliable estimate of the number of occurrence of claims and the total claim amounts is extremely important. Different methods are available in the literature for predicting the claim frequency of a policy for forthcoming years such as Generalized linear models (GLMs), Poisson regression models, Credibility models, Bayesian Models etc. But due to the heterogeneous nature of policies these methods does not produce exact and reliable prediction of future claim frequencies. Besides these conventional statistical methods depends largely on some limiting assumptions such as linearity, normality, independence among predictor variables and a pre-existing functional form relating the criterion variable and predictive variables etc. Recent studies in Artificial Intelligence show that Artificial Neural Networks (ANN) is powerful tools for prediction tasks due to their nonlinear nonparametric adaptive learning properties. In this paper, we developed the procedure for predicting the future claim frequency of insurance portfolio in general insurance using ANN with use of Bayesian credibility inputs with suitable illustration.
Rajitha C. S. and Sakthivel, K. M., “Artificial Intelligence for Estimation of Future Claim Frequency in Non-Life Insurance”, Global Journal of Pure and Applied Mathematics, vol. 13, no. 6, pp. 1701-1710, 2017.