Open AccessDevelopment of Borrowers’ Solvency Assessment Model: Logistic Regression Application
Author(s) -
Dāniels Jukna
Publication year - 2020
Publication title -
journal of economics and management research
Language(s) - English
Resource type - Journals
ISSN - 2255-9000
DOI - 10.22364/jemr.9.02
Subject(s) - solvency , logistic regression , predictability , actuarial science , profit (economics) , econometrics , business , economics , finance , statistics , mathematics , market liquidity , microeconomics
Borrowers’ solvency assessment models can not only increase company’s profit, but also potentially decrease the impact from the negative economic consequences of the crisis. However, there is no consensus on such models. Considering the flaws in the scientific literature, the main aim of this article was to develop the borrowers’ solvency assessment model, which can be applied in practice. The most appropriate method for developing such models was found to be logistic regression, and this research goal is to identify the best modelling approach to achieve the highest borrowers’ solvency predictability. By implementing the best-chosen model, a nonbank lending company could provide a 42.5% lower total borrowers risk of default than without implementing such a model. Depending on the risk policy of the non-bank lending company, three methodologies were developed based on different assumptions about the significance of type 1 error and type 2 error in the company to determine the exact cut-off value