USING MACROECONOMIC INDICATORS TO PREDICT LOAN OUTCOME
Abstract
The Reserve Bank is always concerned about India's loan default ratios, which are constantly on the higher sides. As much as 50% of certain banks' total loans have reached this level. The banking industry has experienced a corresponding increase in credit analytics, which aids the bank in deciding whether to authorise the loan or not, with the latest advancements in data mining. Our methodology may incorporate macroeconomic aspects into the decision-making process using the current credit analytics and, when combined with machine learning techniques, can greatly anticipate how the loan may perform. The lending and borrowing patterns in the country's economy are greatly influenced by variables such as crude oil prices, the dollar exchange rate, monsoon, fiscal deficit, inflation, industrial production, inflation rate, exports and imports, GDP growth, trade deficit, interest rates on government bonds, compensation of public employees, long-term unemployment rate, gross debt, government spending, and consumer spending. By training the proposed model the goal is to reach a practical level of accuracy using methods like KNN, Random Forest, and Logistic Regression. With this, the goal is to produce outcomes that are accurately defined in terms of objective values that are rational and well-suited to loan application criteria, such as "APPROVED" and “DEFAULT”. This method can be used as a tool, to modify the present situation. This novel methodology has a lot of scope for improvements and iterations so that it is finally accepted into the real-world banking sector. This technique has a lot of room for refinement and iteration before it is fully adopted into the real-world banking industry.
Keyword : GDP, KNN Classifier, Loan Default Prediction, Machine Learning, Macro-Economic Parameters, Predicting Loan Outcome
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1] Credit Evaluation Model of Loan Proposals for Indian Banks, Seema Purohit, Anjali Kulkarni 2017 [2] A Machine Learning Approach for Predicting Bank Credit Worthiness, Regina Turkson, Edward Baagyeere, Gideon Wenya 2016 [3] Upadhyay and Roy, Analysis of macroeconomic factors causing non- performing loans in India, International Journal of Business and Globalisation, January 2017. [4] A Comparative Study of Data Mining Methods in Consumer Loans Credit Scoring Management, Wenbing Xiao, Qi Fei, Qian Zhao - 2006 [5] The Impact of Macroeconomic and Bank Specific Factors on Albanian Non-Performing Loans Klejda Gabeshi, European Journal of Sustainable Development Research [6] Laurin, Martynenko and Nilsson, The Influence of Macroeconomic Factors on the Probability of Default - A Study of the Relationship between Default Probabilities and Macroeconomic Variables, August 2009. [7] Machine learning for Credit Scoring: Improving Logistic Regression With Non Linear Decision Tree Effects, Hue Sullivan - 2018 [8] Support Vector Machines for Credit Scoring, Bc. Michal Haltuf-2014 [9] Research and Application of Credit Score Based on Decision Tree Model, Jiang Wei Li 2011 [10] Xiao, Zhaoand Fei, A Comparative Study Of Data Mining Methods In Consumer Loans Credit Scoring Management. J Syst Sci Syst Eng, Dec 2006. [11] Benchmarking state-of-the-art classification algorithms for credit scoring: An update of research, Stefan Lessmann, Hsin-Vonn Seow, Bart Baesens - 2015 [12] Factors Affecting Non-Performing Loans/Assets in the Public Sector Banks of India, Laila Memdani and Shilpam Dubey; April 2018; The Emperical Economic Letters, September 2017 [13] The Relationship between Macroeconomic Factors and the Level of Non-Performing Loans (NPLs) of Commercial Banks in Vietnam, Nguyen Huu Quang , Nguyen Xuan Nhi; March 2017; International Journal of Scientific Engineering Research, Volume 8, Issue 3; ISSN 2229-5518 [14] Muhammad Ishfaq Ahmad, Wang Guohui, Mudassar Hasan, Rizwan Ali , Muhammad Yasir Rafiq , Ramiz ur Rehman; Non-Performing Loans and Economic Growth; October 2016; Scholars Journal of Economics, Business and Management; [15] Abid, Masmoudi and Zouari-Ghorbel, The Consumer Loans Payment Default Predictive Model: an Application of the Logistic Regression and the Discriminant Analysis in a Tunisian Commercial Bank, Springer Science + Business Media New York, May 2016. [16] Purohit, Mahadevan and Kulkarni, Credit Evaluation Model of Loan Proposals for Indian Banks, International Journal of Modeling and Optimization, Vol. 2, No. 4, August 2012. [17] Li,The Hybrid Credit Scoring Model based on KNN Classifier, Sixth International Conference on Fuzzy Systems and Knowledge discovery, 2009. [18] Keramati, Yousefi and Omidvar, Default probability prediction of credit applicants using a new fuzzy KNN method with optimal weight, January 2014. [19] Malekipirbazari and Aksakalli, Risk Assessment in Social Lending via Random Forests, February 2015. [20] Ying, Research on bank credit default prediction based on data mining algorithm, The International Journal of Social Sciences and Humanities Invention 5(06): 4820-4823, 2018. [21] Ghatasheh, Business Analytics using Random Forest Trees for Credit Risk Prediction: A Comparison Study, November 2014. [22] An Artificial Neural Network Approach for Credit Risk Management, Vincenzo Pacelli, Michele Azzolini – 2010 [23] The Consumer Loans Payment Default Predictive Model: an Application of the Logistic Regression and the Discriminant Analysis in a Tunisian Commercial Bank, Lobna Abid Afif Masmoudi Sonia Zouari-Ghorbel-2016 [24] Predicting payments and and Default risks of unsecured consumer loans in online lending, Zhiyong Li, Ke Li, Xiao Yao, Quing Wen-2019 [25] Machine learning approach to drivers of bank lending: evidence from an emerging economy Ozgur et al. Financ Innov (2021)