Effect of Skewness on Optimum Portfolio Selection

In the traditional portfolio theory, we select optimum portfolios by using the first two moments of return distributions, i.e., mean and variance. Traditional portfolio theory assumes that the investors try to maximize the expected return and minimize the variance in portfolio returns. When the returns are not jointly normally distributed (or when the utility function of the investor is not quadratic) the mean-variance efficient portfolio does not maximize the utility of the investor. In addition to mean and variance, the investors also need to consider skewness, the third moment of return distributions. In this paper, we attempt to build an optimum portfolio, keeping in mind the investors’ preference for positive skewness. Using nine years’ monthly returns data for the NSE’s CNX Nifty stocks, we attempt to create portfolios which maximize returns, minimize variance and maximize skewness at the same time. Our results show substantial improvement in (ex post) portfolio performance when we consider skewness in addition to mean and variance.