Socio-economic characteristics of policyholders have a great influence on the probability of insurance policies payments. The statistical data analysis techniques are the most commonly used for the determination of relevant factors and rate of influence. However, the recent wide application have had Data Mining methods, which provide a more accurate assessment in large amounts of information and complex relationships. In this paper it was proposed to use a logistic regression and Bayes classifierfor solving the problem of estimating the probability of insurance payments. It was proposed to use different methods for finding the parameters of the models depending on the type of input data.
private insurance, estimation of probability, logistic regression, Bayes classifier, method of assessingchances, the maximum likelihood method