Loans Portfolio Optimization of Commercial Banks using Genetic Algorithm: A Case Study of Saudi Arabia

Abstract

Click Here:Access Full Text

This study aims at testing the optimal mechanism of bank lending decisions using artificial intelligence techniques. It is based on a sectoral diversification strategy to minimise risk and maximise return of credits facilities portfolio and support bank managers in their decision making. In this context, we suggest a dynamically self-regulating method to optimise the bank lending decisions, by the application of the meta-heuristic approach represented by genetic algorithms optimization. It has been used and improved in more recent empirical studies; the method has become a hot research topic. The reason for choosing GA is its convergence and flexibility in solving multi-objective optimization problems, such as credit assessment, portfolio optimization, and bank lending decision. Furthermore, we have also used Markowitz model to construct a mean-variance optimization problem, based on estimate expected return and risk. Finally, the optimal loans portfolio, among 11 economic activity sectors in the Kingdom of Saudi Arabia during the period 1998-2020, has been selected. We have also compared the results of the genetic algorithm with the classic Markowitz model in its static form.

Keywords: Credit Risks, Optimal Loans Portfolio, Return & Risk, Sectoral Diversification, Genetic Algorithms Optimization

View PDF