WSEAS Transactions on Business and Economics
Print ISSN: 1109-9526, E-ISSN: 2224-2899
Volume 15, 2018
Approximating the Ruin Probability of Finite-Time Surplus Process with Adaptive Moving Total Exponential Least Square
Authors: S. Khotama, S. Boonthiem, W. Klongdee
Abstract: The adaptive moving total least square (AMTLS) method has been used for curve fitting. In this study, AMTLS method is used for the ruin probability fitting and estimation of the ruin probability on an arbitrary initial capital in finite-time surplus process (or risk process). But in reality, it is difficult and complicated to find a fitting method for an appropriate estimate in order to obtain the best performance. So, a new method is developed to estimate the ruin probability of finite-time surplus process. This new method is called adaptive moving total exponential least square (AMTELS) method that applies AMTLS method with least-square fitting exponential. Claim data of motor insurance company from Thailand has used in risk process for the ruin probability fitting. Both AMTLS and AMTELS methods consider weighted function for the distance between node and point with a different constant value . These methods are compared the performance by using the mean squared error (MSE) and the mean absolute error (MAE) that is, the error between the real ruin probability value that is obtained by the explicit formula and the ruin probability fitting value. With these data, the ruin probability approximating examples are given to prove that AMTELS method shows the better performance than AMTLS method. Moreover, AMTELS method with the narrow value shows the better performance than AMTELS method with the wide value.
Keywords: Adaptive moving total least square, exponential claim, least-square fitting exponential, moving total least squares, the ruin probability fitting, weighted function
Pages: 321-328WSEAS Transactions on Business and Economics, ISSN / E-ISSN: 1109-9526 / 2224-2899, Volume 15, 2018, Art. #31