The Mean-Variance Approach

The mean-variance, or risk-return, approach to portfolio analysis is based upon the premise that the investor in allocating his wealth between different assets takes into account, not only the returns expected from alternative portfolio combinations, but also the risk attached to each such holding. This risk is usually assumed to arise out of uncertainty over future asset prices, and can occur in any asset where the expected holding period is less than the term to maturity. Stocks and shares fall into such a category because prices vary according to market conditions, with the risk of capital losses potentially high. However, even the returns from interest-paying capital-certain money deposits may involve an element of uncertainty over the holding period, if the interest rate is subject to market variability. In formal terms, the mean-variance approach assumes that the investor maximises the expected utility obtainable from his portfolio holding, expressed in terms of expected return and risk, subject to the restriction imposed by his budget constraint. Expected return is measured by the mean of the probability distribution of portfolio returns and risk by the standard deviation or variance of the distribution, which provides a measure of the dispersion of possible returns around the mean value. A large standard deviation implies a high probability of big deviations from expected returns, both positive and negative. Such an approach to portfolio analysis stems from Markowitz’s (1952, 1959) studies of efficient portfolio selection and Tobin’s (1958) paper on liquidity preference.