A class of loan portfolio models in fuzzy random environments and hybrid optimization algorithm

Abstract

This paper proposes a class of loan portfolio models in fuzzy random environments based on primitive chance, equilibrium chance, and mean chance, respectively. In the models, the return rates of loan portfolio are all characterized as fuzzy random variables. A hybrid optimization algorithm based on fuzzy random simulation is designed to solve these models. In the algorithm, fuzzy random simulation for estimating the chances is employed to generate a set of input-output data. Then an NN is trained based on the training data to approximate these chance functions. Finally, the NN is embedded in both GA and SPSA. At the end of the paper, an example is given to illustrate the effectiveness of the proposed models and algorithm.