Risk parity approach to asset allocation

Jacob Buhl Jensen

Student thesis: Master thesis

Abstract

Since the nancial crisis, portfolios based on risk diversi cation are of great interest to both academic researchers and market practitioners. They have been employed by several asset mangement rms and their performance appears promising. Since they do not rely on estimates of expected returns, they are assumed to be more robust. Of the multitude of alternative asset allocation portfolios that are being proposed, a collection of approaches broadly referred to under the heading of Risk Parity seems to be gaining traction. This thesis presents a review of asset allocation strategies and develops and backtests these portfolio strategies in a consistent framework. Reviewed are four risk-based portfolios, Global Minimum-Variance, Inverse Volatility, Most Diversi cation Portfolio and Risk Parity. In addition, the traditional 60/40 portfolio is applied, re ecting a portfolio strategy applied by a long-term investor. This thesis uses an evaluation methodology that consider risk adjusted returns, maximum drawdowns, diversi cation ratio, turnover and risk contribution. Based on three emperical backtests some very attractive aspects are to be found in the Risk Parity approach to asset allocation. First, it purports to reduce the dependence on statistical parameters, such as returns, that are di cult to estimate. Second, for a given level of risk, when measured as portfolio volatility, it is much more diversi ed than an any other portfolio strategy. Third, when leverage is applied the risk adjusted performance is superior. However, turnover and associated transaction costs can be a substantial drag on returns. Further, the drawdowns in a Risk Parity strategy can be a signi cant factor when determining ones asset allocation strategy. When backtested on market data, all risk-based portfolios are shown to be e ective in improving portfolio performance over the traditional 60/40 portfolio. Yet, the performance within each risk-based portfolio is mixed. The Inverse Volatility portfolio requires no optimization making it less complex than any of the other portfolio strategies. Nevertheless, the portfolio strategy displays good risk adjusted returns relative to a low turnover. The Global Minimum-Variance portfolio exhibits great risk adjusted returns at the expense of high portfolio turnover. The Most Diversi cation Portfolio entails the highest ex-ante diversi cation but delievers lower risk adjusted returns than the Global-Minimum Variance and at a higher turnover.

EducationsMSc in Accounting, Strategy and Control, (Graduate Programme) Final Thesis
LanguageEnglish
Publication date2013
Number of pages99