[Séminaire Eat and Think] Predicting Financial Distress in Microfinance Institutions & stock market financial contagion

Résumé :
Two research thematic with a common denominator the forecasting of extreme events;

1/ Research axis: Prediction of Microfinance Institution’s financial distress
Problem: MFI’s are threatened by the failure spectrum, even before the advent of the covid pandemic, one of the most famous collapse was the 2010 Andhra Pradesh (India) which was followed by a wave of suicide among them several women’s borrowers unable to repay loans. Despite this crucial situation, the financial literature studies with deep interest the banking & corporate bankruptcy issue and pays little attention to the MFI’s case.

Solution: To set up a credit scoring model able to forecast the MFI’s financial distress relying on internal variables rather than financial variables inherent to the Altman model with a focus on gender variables, i.e, female borrowers and female loan officers borrowers by performing both traditional classifiers, such as logistic regression and machine learning classifiers such as random forest, XgBoost,…

2/ Research axis: Prediction of financial turbulence episodes for stock markets
Problem: Since the great depression, stocks markets were swept away by hurricanes that caused drastic drop in prices named commonly crashes. During the Major crises , among them, Asian crisis, Russian crisis, Internet crisis , .. the extreme stock price volatility in one market spread to other markets/others countries beyond what would be expected. This phenomena is called financial contagion

Solution: To develop an early warning signal able to capture financial contagion and that can be used to build dynamic trading strategies that outperform naïve trading strategy by carrying out rolling principal component analysis.