This is a joint project with Md. Nurujjaman’s group at the National Institute of Technology, India. In the aftermath of stock market crash due to COVID-19, not all sectors recovered in the same way. In , we proposed novel models to capture the different types of recovery profiles for Indian stocks. We also employed the Empirical Mode Decomposition (EMD) for a statistical significance analysis of our model.
So far, the purpose of our models has been descriptive, however prediction of market crash is more crucial and challenging than just observing it. We are now pushing the project in the direction of predicting a future crash in a financial sector—using tools from TDA. Although TDA is not a conventional tool for analyzing non-linear time series, we are emboldened by its recent success in such applications.