Risk parity and other heuristic portfolio allocation strategies

Ing. Daniel Kanyata

Master's thesis

Abstract:

The aim of this thesis is to evaluate the performance of the Risk Parity Portfolio against the 60/40, Equally Weighted and the Global Minimum Variance Portfolios under different market conditions. Two historical securities price datasets, spanning a 17-year period, were used to accomplish this purpose. The first dataset was composed of three indices: A Stock index, Bond index and Commodities index. The second dataset was composed of ten stocks. Monthly and daily logarithmic returns were calculated on the first and second datasets, respectively. A back test was conducted using the rolling window approach to test the performance of the four strategies out of sample. The first dataset was set to a rolling estimation window of 12 months and the second one was set to a three-year rolling estimation window. In both cases, the portfolios were rebalanced and optimized monthly. Four performance metrics were employed: the Sharpe ratio, Maximum Draw Down, Stering ratio and the Omega ratio. The results indicate that when transaction costs are included to mimic a market setup, the Risk Parity Portfolio offers the highest Sterling ratio indicating better performance in market meltdowns and its Sharpe and Omega ratios are second only to the 60/40 portfolio (and its variant). It produces the second least Maximum Draw Down. Otherwise, its performance falls between the Global Minimum Variance Portfolio and the Equally Weighted Portfolio, with respect to the performance metrics used in this thesis.

Abstract:

The aim of this thesis is to evaluate the performance of the Risk Parity Portfolio against the 60/40, Equally Weighted and the Global Minimum Variance Portfolios under different market conditions. Two historical securities price datasets, spanning a 17-year period, were used to accomplish this purpose. The first dataset was composed of three indices: A Stock index, Bond index and Commodities index. The second dataset was composed of ten stocks. Monthly and daily logarithmic returns were calculated on the first and second datasets, respectively. A back test was conducted using the rolling window approach to test the performance of the four strategies out of sample. The first dataset was set to a rolling estimation window of 12 months and the second one was set to a three-year rolling estimation window. In both cases, the portfolios were rebalanced and optimized monthly. Four performance metrics were employed: the Sharpe ratio, Maximum Draw Down, Stering ratio and the Omega ratio. The results indicate that when transaction costs are included to mimic a market setup, the Risk Parity Portfolio offers the highest Sterling ratio indicating better performance in market meltdowns and its Sharpe and Omega ratios are second only to the 60/40 portfolio (and its variant). It produces the second least Maximum Draw Down. Otherwise, its performance falls between the Global Minimum Variance Portfolio and the Equally Weighted Portfolio, with respect to the performance metrics used in this thesis.

Keywords

Risk Parity, Global Minimum Variance, Equally Weighted, 60/40, Backtest, Rolling Window, Portfolio

Language used: English

Date on which the thesis was submitted / produced: 14. 5. 2021

Thesis defence

• Date of defence: 24. 6. 2021
• Supervisor: Ing. Luděk Benada, Ph.D.
• Reader: Ing. Daniel Stašek

Masaryk University

Faculty of Economics and Administration

Master programme / field:
Finance / Finance