Portfolio management of Commodity Trading Advisors with volatility-targeting

Marat Molyboga

Abstract

This paper shows analytically that a volatility-targeted allocation methodology improves the risk-adjusted performance of portfolios under a broad set of assumptions regarding the serial correlation of returns and the dependence of the expected Sharpe ratio on the level of volatility. The author examines the impact of volatility-targeting on portfolios of Commodity Trading Advisors within the large-scale simulation framework of Molyboga and L’Ahelec, which accounts for the real-world constraints on institutional investors. They find a consistent and statistically significant improvement in the out- of-sample returns that ranges between 0.53% and 0.80% per annum, on average. The performance enhancement is robust to portfolio size and manager selection, and it is implementable inside managed account investments.