Detecting Market Regime Changes

Keywords: regime modeling, hidden Markov model, naive Bayes classifier, logistic regression, support vector machine

This project was done for the requirements of the class 46-927 Machine Learning II, along with my group members Dhruv Baid, Rohan Prasad, and Sarthak Vishnoi. It can be found on GitHub.

Used hidden Markov models to label regimes on S&P 500 data, then trained three classifiers (naive Bayes, logistic regression, and support vector machines) to predict regimes with the intent of improving an existing trading strategy.