Two approaches to calculating correlated reserve indications across multiple lines of business

Document description

As reserving actuaries focus more on reserve ranges and less on point estimates, the question of how to develop a reasonable reserve range in the aggregate becomes more and more relevant. When working with a single set of 'best estimates', the answer is simple - assuming all the best estimates are the mean values for each block of business being analyzed, the best estimate for the total is equal to the sum of the parts. However, if the by line best estimates are other than the mean values, the sum of the parts is not the same as the best estimate for the aggregate. This paper presents two possible approaches to developing aggregate reserve indications when looking at results other than the mean value. The approaches both rely on a simulation model. One takes in the actuary's judgment as to the correlations between the different underlying blocks of business and the second uses bootstrapping to eliminate the need for the actuary to make judgment calls about the nature of the correlations.