Multi-Dimensional Risk and Mean-Kurtosis Portfolio Optimization

Journal of Financial Management and Analysis, Vol. 21, No. 2, December 2008

Posted: 7 May 2009

See all articles by James Stacey

James Stacey

Memorial University of Newfoundland (MNU) - Faculty of Business Administration

Date Written: March 24, 2009

Abstract

The mean-variance portfolio optimization theory of Markowitz assumes that stock returns are distributed according to normal probability density functions (pdfs). In reality, stock returns are more accurately described by leptokurtic pdfs which have kurtosis greater than zero. Stocks with leptokurtic distributions of returns are conventionally considered to be inherently more risky than stocks with normal pdfs. This paper examines portfolio optimization using the kurtosis as a risk measure. Maximizing the kurtosis as a function of portfolio weights is equivalent to maximizing the probability of large fluctuations from the mean which is counter-intuitive and contrary to normal practice. It is argued that risk is multidimensional and that the kurtosis is an ambiguous multi-dimensional risk measure.

Keywords: Portfolio, Optimization, Kurtosis, Riskmeasure

JEL Classification: C12, C13, C61, C87, G32, N20

Suggested Citation

Stacey, James, Multi-Dimensional Risk and Mean-Kurtosis Portfolio Optimization (March 24, 2009). Journal of Financial Management and Analysis, Vol. 21, No. 2, December 2008, Available at SSRN: https://ssrn.com/abstract=1400685

James Stacey (Contact Author)

Memorial University of Newfoundland (MNU) - Faculty of Business Administration ( email )

St. John's, Newfoundland A1B 3X5
Canada

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
826
PlumX Metrics