Improvement in Bank Financing Access for Peruvian MSE using prediction and classification models

Authors

Samaniego Osorio, Alvaro Danilo

Rojas Polo, Jonatán Edward

Cáceres Cansaya, Alexia Andrea

Gilardino Arias, Alessandro Oscar

Viamonte Yucra, Felipe

Benavente Sotelo, Renzo Alejandro

Abstract

Economic development in Peru has a directly proportional relationship with the growth of micro and small enterprises (MSE), being this business the ones responsible for work generation and poverty reduction. As stated by SUNAT (the national superintendence for taxes), in this decade MSE represent 42% of the GDP; in addition, they formed 98.6% of the established businesses and about 76% of the economically active population. However, a high number of MSE are still facing obstacles that reduce their mid and long-term survival rate. Some MSE develops productive process know-how but this does not guarantee their growth because of a group of factors that need to be taken into account. Currently, the average survival years of a micro and small enterprise are 6.2 and 8.3 years, respectively, and the business mortality is 6 of every 100 MSE, mainly caused by operational and financial weaknesses, such as funding access. The current research starts with the difficulties that MSE face when they want to get funds from a financial institution. A dedicated study was made about the different variables that are relevant to the approval or rejection of a bank-issued credit. Data mining techniques, such as predictive models construction and logistic regression were considered encountering behavioral patterns from several institutions looking to strengthen the MSE access for financing.