A DYNAMIC SELECT SECTOR SPDRS ETFS PORTFOLIO OPTIMIZATION MODEL WITH REGIME-SWITCHING ECONOMIC INDICATORS

Abstract

This thesis studies a dynamic Select Sector SPDRs ETFs portfolio optimization problem. The objective of the optimization model is to maximize the risk-adjusted expected return of a portfolio similar to a logarithmic utility maximization. The conditional value-at-risk measure is chosen to be an additional risk exposure constraint. The vector auto-regression (1) regime-switching economic factor model estimated with the expectation-maximization algorithm is employed to identify different market regimes over time. The expected ETFs returns and their variance-covariance matrix used in the objective function of the optimization model are generated by a regime-switching asset pricing model. Both regime-switching models have proven to be superior to respective single-regime models due to their greater predictive ability. The optimized portfolio performance evaluated by Sharpe ratio, Treynor ratio and Jensen’s alpha are all statistically significant compared to those of the equally weighted ETFs portfolio and S&P 500 stock index. This illustrates that incorporating the regime-switching technique, the portfolio optimization model is effective and successful under both bull and bear market conditions.