Loan Portfolio as an Object of Agent-Oriented Modeling

Abstract

The microlending market currently has certain specific features, including the fact that borrower behavior influenced by certain external challenges. Aim. This study aims to find a solution to the problem of portfolio management for microlenders by determining common patterns in the behavior of small and medium enterprises (SMEs) under certain conditions considering the projection of these patterns on a microlender’s portfolio. Methods. This study uses system analysis, logic analysis, and simulation modeling. Results. We propose an approach for solving the problems of portfolio management for microlenders who provide loans to SMEs. We analyze modern approaches to the development of simulation models that would adequately reflect the factors and processes influencing the relationship between microlenders and SMEs. We conclude that an agent-oriented model of active borrower behavior should be developed using AnyLogic based on a two-dimensional classification of active borrowers. The optimal spatial detail of the model’s environment (grid) is being discussed.