Multi–objective Evolutionary Algorithms for the Risk–return Trade–off in Bank Loan Management

Abstract

Multi-criteria decision-making is an increasingly accepted tool for decision-making in management. In this work, we highlight the application of a novel multi-objective evolutionary algorithm, NSGA-II, to the risk-return tradeoff for a bank-loan portfolio manager. The manager of a bank operating in a competitive environment faces the standard goal of maximizing shareholder wealth. Specifically, this attempts to maximize the net worth of the bank, which in turn involves maximizing the net interest margin of the bank (among other factors, such as noninterest income). At the same time, there are significant regulatory constraints placed on the bank, such as the maintenance of adequate capital, interest-rate risk exposure, etc. The genetic algorithm-based technique used here obtains an approximation to the set of Pareto-optimal solutions which increases the decision flexibility available to the bank manager, and provides a visualization tool for one of the trade-offs involved. The algorithm is also computationally efficient and is contrasted with a traditional multi-objective function -the epsilonconstraint method.

A self-adaptive linear evolutionary algorithm for solving constrained optimization problems

“…There exists some disconnected spaces on Pareto front in Fig.3. The Pareto front consists of 3,5], and x * 2 = 3. For case 2, constraint conditions are also two inequality statement.…”

Multi-criteria decision-making model for whole life costing design

“…The fuzzy multi-attribute decision-making (MADM) method has been used to support the decisionmaking process in sustainable housing (Li and Shen 2002). Mukerjee (2002) used a multi-objective decisionmaking method to solve optimisation problems in riskreturn trade-off in bank loan management.…”

Evolutionary Optimization for Multiobjective Portfolio Selection under Markowitz’s Model with Application to the Caracas Stock Exchange

“…They demonstrated that in this case, the selection of the portfolios becomes more complicated because the problem model has to manage mixed integer variables and nonlinear objectives. From a more general point of view, Mukerjee et al [20] utilize NSGA-II to implement a decision-making multicriteria model used in the risk/performance negotiation by a bank loan manager. Two models with respect to this negotiation were considered.…”