Evaluating the Empirical Performance of Risk-based Portfolio Strategies in the Korean Stock Market

Park, Soonchae; Eom, Young Ho; Hahn, Jaehoon

Korean Journal of Financial Studies 2016;45(2):247-284.
Published online April 30, 2016.

Soonchae Park, Young Ho Eom, Jaehoon Hahn

위험기반 포트폴리오 전략의 성과에 관한 실증 연구

박순채, 엄영호, 한재훈

Abstract

This paper is an empirical investigation of the performance of various portfolio selection strategies applied to the Korean stock market data. The portfolio strategies considered include the classical mean-variance optimal portfolio of Markowitz (1952), various extensions of the Markowitz model designed to reduce estimation error, and several risk-based portfolios such as equal-weighted risk contribution portfolio and most-diversified portfolio. We evaluate the out-of-sample performance of various portfolio strategies relative to the equal-weighted (or 1/N) portfolio as a benchmark, using six different sets of portfolios as the investment opportunity set. We find that risk-based portfolio strategies perform better than the classical mean-variance optimal portfolio or its various extensions in terms of the Sharpe ratio, certainty-equivalent return, or turnover. When compared against the benchmark 1/N portfolio strategy, risk-based portfolio strategies show better performance, although the difference is statistically insignificant in many cases. These main findings are also robust to varying the estimation period and portfolio rebalancing frequency.

Key Words: 성과지표,위험기반 포트폴리오,추정위험,평균-분산최적화,포트폴리오 투자전략,Estimation Risk,Mean-Variance Optimization,Performance Evaluation,Portfolio Strategies,Risk-based Portfolio