Dr. Sotirios Chatzis published a new paper in  IEEE Transactions on Pattern Analysis and Machine Intelligence, the most prestigious journal in the fields of Machine Learning and Pattern Recognition, and the highest impact factor journal in Electrical and Computer Engineering.

In his new article entitled “Gaussian Process-Mixture Conditional Heteroscedasticity,” Dr. Chatzis introduces a new method based on tools from Bayesian non-parametrics for modeling and predicting short-term volatility in financial markets. The proposed method has outperformed the seminal GARCH method (and some of its recent extensions), a mainstay in financial engineering introduced two decades ago by Nobel Laureate R. F. Engle.