Robust Portfolio Optimization

Financial investment is a crucial activity for both institutions and individuals. In particular, the construction of a portfolio of financial assets is at the core of financial investment decision making. Effective portfolio management enables many benefits such as better pensions and funds for education. The essence of portfolio construction is to balance expected returns (i.e. financial benefit) and risk (i.e. the possibility of financial loss). Quantitative methods play a crucial role in the generation of portfolios that optimizes the trade-off between risk and reward. However, most methods need estimates of important parameters such as future expected returns and other statistical quantities that go into the calculation of risk. A major limitation is that most models are not able to effectively incorporate that fact that these estimates change over time (e.g. due to unexpected turbulence in the markets) thus making previous estimates obsolete or misleading resulting in overly risky or underperforming portfolios. The research in this proposal aims to effectively mitigate parameter uncertainty by considering robust and dynamic models for portfolio optimization. As our partner organization Manulife Financial routinely constructs financial portfolios the research from this internship would be greatly beneficial for Manulife Financial.

Faculty Supervisor:

Roy Kwon

Student:

Partner:

Manulife Financial

Discipline:

Engineering

Sector:

Finance and Insurance

University:

University of Toronto

Program: