+44 1865 273524
University of Oxford
Andrew Wiles Building
Radcliffe Observatory Quarter
I study functionally-generated portfolios in the context of Stochastic Portfolio Theory (SPT), a robust framework which makes pathwise comparisons between the performance of such portfolios and a market benchmark. The main strength of this area is that it does not require any model postulation or calibration, and allows for and exploits arbitrages. In particular, my research focuses on:
the construction and theoretical verification of functionally-generated portfolios, and their implementation using real-world data and machine learning techniques;
relating functionally-generated portfolios to optimal transport theory;
applying stochastic control techniques through a hedging interpretation of optimal relative arbitrage;
incorporating the possibility of defaults by extending the pathwise decomposition of market-relative performance, and by using results from extreme value theory.
I am also a member of the Oxford-Man Institute of Quantitative Finance.
Major / Recent Publications:
- with Y.-L. Kom Samo. Stochastic Portfolio Theory: A Machine Learning Perspective, to appear in UAI 2016 conference proceedings. Also on arXiv, May 2016.
- with I. Karatzas. Diversity-Weighted Portfolios with Negative Parameter, Annals of Finance, July 2015. Also on arXiv, April 2015.
- Topics in Stochastic Portfolio Theory, first-year transfer thesis, October 2014.