Aggregation of Correlated Risk Portfolios: Models and Algorithms, [Discussion]

In response to a request for proposal from the Committee on the Theory of Risk, Shaun Wang has written a paper that significantly advances, to quote the proposal, “the development of tools and models that improve the accuracy of the estimation of aggregate loss distributions for blocks of insurance risks.” Dr. Wang’s charge was to “assume a book of business is the union of disjoint classes of business each of which has an aggregate distribution...The classes of business are NOT independent...The problem is how do you calculate the aggregate distribution for the whole book.” Dr. Wang’s paper covers a variety of dependency models and computational methods. This discussion of his paper delves more deeply into a particular dependency model—correlation caused by parameter uncertainty—and then shows how his work applies to calculating the aggregate loss distribution for this case with one particular computational method— Fourier Inversion.