Abstract: Using the example of a practical use case in quantitative finance, this workshop will look at how recent developments in automated machine learning and interpretability can help quantitative and fundamental investors build, test and understand powerful AI models that support their investment process. Ayub and Peter will use data from J. P. Morgan and automated machine learning from DataRobot to illustrate an end-to-end iterative workflow to analyse the factors that drive stock performance following a cut in dividend expectations and build a set of actionable models that may be incorporated into an investment process. They will contrast this approach with "traditional" quantitative research and also identify and address some common misconceptions and mistakes made in AI-driven investing.
Bio: Ayub Hanif, a Vice President, joined J.P. Morgan’s Global Quantitative and Derivatives Strategy team in 2014. The team consistently ranks in Institutional Investor’s top-3. Before switching to the Quant group, Dr. Hanif developed technical strategy for Equity Risk & Structuring within Equity Derivatives Trading. Prior to joining J.P. Morgan, he completed his PhD in machine learning and computational finance from University College London and Harvard University. After completing an MSci in Computer Science from Queen Mary, University of London, Dr. Hanif worked for two years in market intelligence for Merrill Lynch Commodities Analytics. He also holds a Master of Research degree in Financial Computing from University College London. He has developed a number of artificial intelligence models in applied mathematics, computational finance and astrophysics, and has authored top-tier publications in trading, computational finance, machine learning, natural computation and astrophysics.