Factor-based Robust Index Tracking | Hanlon Financial Systems Center

Factor-based Robust Index Tracking

Factor-based Robust Index Tracking

seminar date: 
Thursday, April 14, 2016 - 5:45pm
seminar location: 
Roy Kwon, University of Toronto

The construction of portfolios that mimic the risk-return characteristics of known market indices such as the S&P 60/100/500 have taken on increased importance as ETFs on market indices have proliferated.

Fundamental to this construction is the fact that full replication (i.e. holding a portfolio of exactly the same stocks as an index in the same proportion at all times) is not practical due to overriding transaction costs.

In this talk we consider employing a strict subset of assets selected from the market index such that the expected return is maximized subject to both risk and tracking error constraints. The model is in its nominal form is an extended mean-variance type formulation, thus the model will be prone to the adverse effects of estimation error of parameters.  

To mitigate noise from parameter estimation we propose a robust optimization approach for tracking a market index. Parameters are generated from a robust version of the Fama-French 3 factor model whereby uncertainty sets for the expected return and factor loading matrix are generated.  The resulting model is a mixed integer second-order conic optimization problem.

Computational results in tracking the S&P 100 and S&P 500 show that the robust model can generate tracking portfolios that have better tracking error and Sharpe ratio than those by nominal models.    




Roy H. Kwon is associate professor of operations research in the Department of Mechanical and Industrial Engineering at the University of Toronto (U of T). He also holds an appointment in the Masters of Mathematical Finance (MMF)  program at the U of T where he has been teaching a course in operations research for finance for the last 13 years. 

He holds a Bachelors degree in mathematics from the University of Chicago, a Masters degree in operations research from the University of Michigan, and a PhD degree in electrical and systems engineering from the University of Pennsylvania in 2002 where shortly after he joined the University of Toronto.

The focus of his research has been on financial optimization and risk management. Current projects focus on robust and distribution-free approaches in portfolio asset allocation and option risk management. Much of this research has been supported by major Canadian banks and financial institutions such as the Royal Bank of Canada (RBC), the Bank of Montreal (BMO), and Manulife, Inc. 

Prof. Kwon is also the author of the book “Introduction to Linear Optimization and Extensions with MATLAB” published by CRC Press and serves as an associate editor for the journal Optimization and Engineering.

Last but not least he has been the director of an honours undergraduate engineering program in quantitative finance and operations research at the U of T.