Genetic algorithm based model for optimizing bank lending decisions
Introduction
It is clear that the financial crisis was accompanied by a reduction in the credit supply available to all customers (Aguilar, Valenzuela, & Ortiz, 2015). According to recent studies (Judit, Wang, 2012, Michael, Rohwedder, 2010), the nature of the latest financial crisis in US has brought to the fore concerns regarding bank’s ability to continue with its traditional bank lending strategies. The main channel through which a banking crisis affect the real economy relates to its ability to provide the credit needed given credit constraints imposed on them. Hence, a key question at the heart of the any financial crisis is whether and how did the banking sector managed to distribute the limited credit available in a way that maximizes their profits in the time of crisis (Manju, Rocholl, & Steffen, 2011). Therefore, there is a need to set an optimal mechanism of bank lending decisions that will maximize the bank profit in a timely manner. The inability of banks to manage loan portfolio efficiently may result in a credit crunch. A credit crunch is often caused by a sustained period of careless and inappropriate lending, resulting in losses for lending institutions and investors in debt when the loans turn sour and the full extent of bad debts becomes known. These challenges have led to a rise in more formal and accurate methods to optimize the lending decision and minimize loan risks. In conjunction, bank lending decision has become a primary tool for financial institutions to increase profit, reduce possible risks, and make managerial decisions.
Unlike the traditional statistical and mathematical programming techniques such as discriminant analysis, linear and logistic regression, linear and quadratic programming, or decision trees; intelligent systems have proven their ability to overcome different challenges in financial research areas, specially in banking loan portfolio optimization (Eletter, Yaseen, Elrefae, 2010, Ghodselahi, Amirmadhi, 2011, Marque, Garci, Nchez, 2013, Nazari, Alidadi, 2013). However, most of them are focused on either credit scoring - to determine whether the applied customer is eligible to get the required loan - (Ghodselahi, Amirmadhi, 2011, Marque, Garci, Nchez, 2013), or portfolio selection (Berutich, Francisco, Luna, Quintana, 2016, Lwin, Qu, MacCarthy, 2017, Saborido, Ruiz, Bermúdez, Vercher, Luque, 2016) aiming to choose the optimum stocks that maximize the customer profit.
Thus, the problem of bank lending decision in a credit crunch environment- where all applicable customers are eligible to get the desired loan - is an NP-hard optimization problem that can be solved using meta-heuristic algorithms such as evolutionary algorithms (Bhargava, 2013, Ghosh, Tsutsui, 2003, Metawa, Elhoseny, Kabir Hassan, Hassanien, 2016). The evolutionary computing methods are highly capable of extracting meaning from imprecise data and detecting trends that are too complex to be discovered by either humans or other conventional techniques. For that purpose, this paper proposes an intelligent model based on Genetic Algorithm (GA) to organize bank lending decision in a highly competing environment with credit crunch constraint.
The main contribution of this paper is the creation of a GA model that facilitates how banks would make an efficient decision in case of a cut back on lending supply when faced with a liquidity shock, while staying focus on the main objective of bank profit maximization. The main focus of the GA model is two-fold: to stabilize systemically banks while achieving maximum profit, and to establish the capital base so that banks would increase lending efficiently. The GA model takes into account the margins along which banks adjust their loan portfolios in response to the crisis, either through imposing credit supply contractions across all types of borrowers, or disproportionally cut back credit for some specific types of loans. Multiple factors including loan characteristics, creditor ratings and expected loan loss are integrated to GA chromosomes and validation is performed to ensure the optimal decision.
The remaining of this paper is organized as the following: Section 2 discusses the literature review of the bank lending decision’s problem and the recent work to solve this problem using GA. Section 3 describes GAMCC and the GA data representation. Then, Section 4 discuss the experimental results. Finally, Section 5 concludes the paper.
Section snippets
Literature review
The optimal loan portfolio selection requires a solution of a high-dimensional nonlinear program and is computationally challenging (Elhachlouf, Guennoun, Hamza, 2012, Sefiane, Benbouziane, 2012). Loan portfolios commonly have a large number of different loans ranging from hundreds to thousands of loans (Melennec, 2000). Every loan is characterized by a highly dimensional vector of loan characteristics including; credit score, collateral, interest rate, loan balance, loan purpose, loan age and
Problem representation
This paper proposes an efficient, GA-based model to maximize bank profit in lending decision. In our proposed method, the lending decision is dynamically decided based on customer’s loan characteristics. With the assumption that all customers are applicable (good) to get the required loan, GA is employed to search for the most suitable customers depending on a set of factors such as loan age, loan size, loan interest rate, loan type, and borrower credit rating. A set of other depending
Results and discussion
First, all data are obtained from World Bank public database for year 2016. In our evaluation, the GA parameters used in our experiments are listed in Table 5. The analysis of the results is conducted based on 10 experiments.
To evaluate GAMCC, two scenarios are created to study the routine method performance in real and simulated environments. In the simulation scenario, GAMCC randomly generates customer loan’s characteristics for a defined number of customers with different values and exports
Conclusion and future work
The bank lending decision problem in a credit crunch environment is an NP-hard optimization problem that can be solved using meta-heuristic algorithms such as evolutionary algorithms. To avoid the complexity and time consumption of the traditional statistical and mathematical programming techniques such as discriminant analysis, linear and logistic regression, linear and quadratic programming, or decision trees; intelligent techniques have proven their ability to overcome the different
References (35)
- et al.
Robust technical trading strategies using GP for algorithmic portfolio selection
Expert Systems with Applications
(2016) - et al.
A literature review on the application of evolutionary computing to credit scoring
Journal of the Operational Research Society, 2013
(2013) - et al.
Portfolio optimization of commercial banks-an application of genetic algorithm
European Journal of Business and Management
(2013) - et al.
Portfolio selection using genetic algorithm
Journal of Applied Finance and Banking
(2012) - et al.
Investment and bank credit recovery after banking crises
Journal of Financial Stability
(2016) - et al.
Optimization of bank loan portfolio management using goal programing technique
International Journal of Research in Applied Natural and Social Sciences
(2014) - et al.
Genetic algorithms and Darwinian approaches in financial applications
Expert Systems with Applications
(2015) Corporate bond and commercial loan portfolio analysis
(1996)A note on evolutionary algorithms and its applications
Adults Learning Mathematics: An International Journal
(2013)- et al.
Does credit crunch investment down? New evidence on the real effects of the bank-lending channel
Review of Financial Studies
(2016)
Improving loan portfolio optimization by importance sampling techniques: Evidence on Italian banking books
Economic Notes by Banca Monte dei Paschi di Siena SpA
Risk overhang and loan portfolio decisions: Small business loan supply before and during the financial crisis
The Journal Of Finance
Neuro-based artificial intelligence model for loan decisions
American Journal of Economics and Business Administration
Stocks portfolio optimization using classification and genetic algorithms
Applied Mathematical Sciences
Balancing energy consumption in heterogeneous wireless sensor networks using genetic algorithm
IEEE Communications Letters
Application of artificial intelligence techniques for credit risk evaluation
International Journal of Modeling and Optimization
Advances in evolutionary computing: Theory and applications
Cited by (223)
Partial and cost-minimized computation offloading in hybrid edge and cloud systems
2024, Expert Systems with ApplicationsHuman-AI collaboration to mitigate decision noise in financial underwriting: A study on FinTech innovation in a lending firm
2024, International Review of Financial AnalysisA step-size follow-the-leader optimization algorithm with an improved step parameters
2023, Decision Analytics JournalOperational research and artificial intelligence methods in banking
2023, European Journal of Operational ResearchOptimal cost-sensitive credit scoring using a new hybrid performance metric
2023, Expert Systems with Applications