Improved Forecasts of Higher-Order Co-moments and Implications for Portfolio Selection

In the presence of non-normally distributed asset returns, optimal portfolio selection techniques require not only estimates of variance-covariance parameters, but also estimates of higher-order moments and comoments of the return distribution.

Author(s):

Lionel Martellini

Professor of Finance and Scientific Director of the EDHEC Risk and Asset Management Research Centre

Volker Ziemann

PhD, Research Engineer at the EDHEC Risk and Asset Management Research Centre

This paper expands on existing literature, which has focused mostly on the covariance matrix, by introducing improved estimators for the coskewness and cokurtosis parameters. In an empirical analysis, we find that the use of these enhanced estimates leads to significantly better out-of-sample performance.
Pdf
Improved Forecasts of Higher-Order Co-moments and Implications for Portfolio Sel...
(-1.00 B)
Type: Working paper
Date: le 29/11/2007
Extra information : Pour plus d'informations, nous vous prions de vous adresser à Joanne Finlay, Direction de la recherche de l'EDHEC [ joanne.finlay@edhec.edu] Les opinions exprimées sont celles de l'auteur et n'engagent pas la responsabilité de l'EDHEC.
Research Cluster : Finance

See Also

Coursera: Global Financing Solutions by EDHEC & Société Générale
News
- 21-03-2019
Before the launch in june of a special track in Data Science and Machine Learning in...
How To: 8 tips for choosing your MBA
News
- 08-03-2019
Embarking on an MBA can be life-changing. It’s also an investment on many levels; you’...
European Excellence and Make an Impact Scholarships available for September 2019 Intake
News
- 07-03-2019
Two New Scholarship Programmes for 2019 EDHEC Business School has added two new Global...
Paper co-authored by Lionel Martellini and Vincent Milhau wins the 20th Annual Bernstein Fabozzi/Jacobs Levy Award for Outstanding Article from the Journal of Portfolio Management
News
- 04-03-2019
The paper, entitled "Proverbial Baskets Are Uncorrelated Risk Factors! A Factor-Based...