By: Dr. Brian Thompson
There are multiple objectives for portfolio risk measurement. Current portfolio return and risk should be appraised on an ongoing basis to make certain that client guidelines are followed and that performance results were achieved at acceptable risk levels. In a comparative multi-manager analysis, it is also important to evaluate performance while considering the level of risk taken. It is also important to know how closely a portfolio tracks the benchmark and how successful a manager is in exceeding the benchmark in both up and down markets. A range of appropriate tools are needed to compare managers on a risk/return basis against benchmarks and relevant peers.
Multiple Measurement Solutions
First Rate has developed a broad set of risk metrics and performance adjusted risk measures to appraise manager risk on both absolute and comparative bases. The risk measures can be classified into four basic groups:
A. Volatility and Downside Risk - The standard deviation of returns is provided as a well accepted measure of return volatility when the returns are normally distributed. This measure describes the variability of returns as symmetric both above and below the mean return. However, the standard deviation is not the best measure of return variability when the returns are skewed or otherwise non-normal.
First Rate offers two downside risk measures to represent risk in situations where returns are skewed. The first measure is the downside standard deviation, which is also known as the semi-deviation. This measure focuses on the variability of returns below the mean return. The downside deviation is the second downside risk measure and focuses on the variability of returns that fall below a return objective or minimum acceptable return (MAR). A partial listing of potential target or MAR returns includes the following: (1) zero, where any market loss is considered not acceptable; (2) the actuarial return assumption used to project liabilities; (3) the risk-free rate such as the 90-day treasury return; and (4) a specified benchmark return.
B. Capital Market Measures -The basic capital market measures such as beta and residual risk can be used to decompose portfolio risk into systematic and non-systematic risk. Systematic risk is non-diversifiable and can be estimated as the risk explained by the Capital Asset Pricing Model. The residual risk is the non-systematic or diversifiable risk. The residual risk measures diversification as volatility of the random error term around the regression line. The alpha is also provided as a measure of manager skill. In addition, the R-squared or coefficient of determination is provided that also measures diversification as the fraction of the portfolio variance that is explained by the regression line.
C. Risk-Adjusted Returns - Five risk-adjusted return measures are provided. Four of these measures are developed in terms of excess return per unit of risk, where the term definitions vary from measure to measure. The fifth measure is the M-squared measure, which is an adjustment of the Sharpe ratio that results in a performance measure for the portfolio at a risk level scaled to the volatility of the benchmark.
Back