Metaheuristic-based portfolio optimization in peer-to-peer lending platforms

• Published in: International journal of system assurance engineering and management 2024-08, Vol.15 (8), p.3629-3642
• Main Authors: Abbasi, Hadis, Bamdad, Shahrooz, Rahimi, Morteza
• Format: Article
• Language: English
• Subjects: Engineering; Engineering Economics; Logistics; Marketing; Organization; Original Article; Quality Control; Reliability; Safety and Risk
• ISSN: 0975-6809; 0976-4348
• DOI: 10.1007/s13198-023-02074-0

In recent years, researchers have paid increasing attention to Peer to Peer lending market. In this lending method, the lenders don’t usually have enough knowledge to understand how much money must be allocated in each loan. So, it is important to help investors to assess the loans. According to the needs of investors, generally, a bi-objective optimization model is considered that maximizes return and minimizes risk. In current work, internal rate of return  as return and probability of default as risk variable is considered. First, an artificial neural network is used to evaluate the return. Then, we apply logistic regression to predict the probability of loan default. After providing the model, due to its complexity, one of the essential issues is solving it. In this study, the model is solved by metaheuristic algorithms. This is done from different points of view, including non-dominated sorting genetic algorithm II, multi-objective particle swarm optimization, and Pareto envelope-based selection algorithm II. Finally, from the perspective of an investor, the best algorithm is introduced.