Narrow Results

Machine learning has successful applications in credit risk management, portfolio management, automatic trading, and fraud detection, to name a few, in the domain of finance technology. Reformulating and solving these topics adequately and accurately is problem specific and challenging along with the availability of complex and voluminous data. In credit risk management, one major problem is to predict the default of credit card holders using real dataset. We review five machine learning methods: the $k$-nearest neighbors decision trees, boosting, support vector machine, and neural networks, and apply them to the above problem. In addition, we give explicit Python scripts to conduct analysis using a dataset of 29,999 instances with 23 features collected from a major bank in Taiwan, downloadable in the UC Irvine Machine Learning Repository. We show that the decision tree performs best among others in terms of validation curves.

Credit scoring is used by institutions in managing the credit risk associated with their borrowers. In addition to the traditional variables used in credit scoring, recent developments in machine learning have allowed the inclusion of data sources, such as ongoing payment behaviors and patterns, creating more robust and dynamic models. While such studies have more commonly been done for credit card data, there is limited research in the context of microfinance. Investigating this niche, this study focuses on uncovering how delinquency can be characterized and whether delinquency states exist within microfinance loans using a dataset from a microfinance institution (MFI) in the Philippines. Further, Markov chains and transition state matrices from the dataset were used to predict repayment sequences, and factors such as borrower characteristics, loan information and repayment behaviors were used in predicting the final outcome of the loans. The findings show that Markov chains can predict repayment sequences using a transition state matrix based on the full term of the loan. This model outperformed other models created from transition state matrices based on the first 8, 12, and 17 weeks of the loan term, indicating the repayment performance in the first weeks is not representative of the full term and cannot be used to accurately predict future repayment patterns. The equivalent second-order Markov chains were also examined which resulted in better predictions generally. Meanwhile, in predicting the final outcome of the loans, random forest models outperformed decision tree models using an accuracy and modified accuracy metric for evaluation. Across all models assessed in this study, behavioral characteristics and payment patterns consistently fared higher in terms of feature importance than borrower and loan characteristics. Particular to tiering delinquency states, the combination of average days overdue and delinquency streak appeared to be fitting resulting in delinquency tiers Onset Delinquency and Significant Delinquency. The ability to predict repayments aims to provide MFIs with better oversight of their loan portfolios. With an emphasis on the early detection of delinquency, predictive models using relevant features, such as those identified in this study, may allow the implementation mitigating actions at specific time periods to improve overall repayment rates among borrowers of MFIs.

This paper uses loan-level data from Thailand's National Credit Bureau to study household debt over the life cycle of borrowers. We decompose two aggregate and commonly used measures of debt—debt per capita and delinquency rate—into components that unveil the extensive and intensive margins of household indebtedness. We find a striking inverted-U life-cycle pattern of indebtedness as predicted by economic theories. However, peaks are reached at different ages for different loan products and different lenders. We also find that debt has expanded over time for all age groups. Younger cohorts seem to originate debt earlier in their lives than older generations. Meanwhile, older borrowers remain indebted well past their retirement age. Finally, we find a downward pattern of delinquency over the life cycle. Our findings have important policy implications on financial access and distress of households as well as on economic development and financial stability of the economy.

Machine learning has successful applications in credit risk management, portfolio management, automatic trading, and fraud detection, to name a few, in the domain of financial technology. Reformulating and solving these topics adequately and accurately is problem specific and challenging along with the availability of complex and voluminous data. In credit risk management, one major problem is to predict the default of credit card holders using real data set. We review five machine learning methods: the k-nearest neighbors, decision trees, boosting, support vector machine, and neural networks, and apply them to the above problem. In addition, we give explicit Python scripts to conduct analysis using a data set of 29,999 instances with 23 features collected from a major bank in Taiwan, downloadable in the UC Irvine Machine Learning Repository. We show that the decision tree performs best among others in terms of validation curves.