Efficient Grouping of Policies with Machine Learning and Artificial Intelligence
Every actuary working with the stochastic cash flow models in life insurance knows how inefficient (and from an actuarial point of view, unsatisfying) the waiting for the results of a stochastic run can be. Moreover, the IT and cost controlling departments are affected by the high expenses incurred by these runs.
The main culprit for the long runtimes of cash flow models is the large number of insurance contracts. At the same time new actuarial and analytical methodologies offer clever ways to reduce the number of contracts that need to be projected and hence reduce the model runtime.
In his talk, Zoran will present a new solution which combines certain analytical methodologies such as linear programming with machine learning and artificial intelligence leading to compression levels exceeding everything that has so far been seen within the insurance industry.