An Empirical Evaluation of Sensitivity Bounds for Mean-Variance Portfolio Optimisation

Andrew Paskaramoorthy and Matthew Woolway

Finance Research Letters, 2022, vol. 44, issue C

Abstract: It is commonly thought that a poorly conditioned covariance matrix causes the sensitivity of mean-variance optimised portfolios to deviations in expected return forecasts. In this research, we question this explanation and show that it does not necessarily hold when a budget constraint is included in the optimisation problem. Our research is centred on the analytical results derived by Best and Grauer (1991) that describes the maximum amount by which a portfolio and its performance can change due to changes in the mean vector. Our empirical analysis shows that these derived bounds can overstate the actual corresponding maximums by several orders of magnitude. We explain these results with reference to the original derivations. In conclusion, we find that these bounds, and the condition number, in particular, are unable to characterise portfolio sensitivity.

Keywords: portfolio optimisation; error-maximisation; portfolio sensitivity; ill-conditioning (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S154461232100146X
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finlet:v:44:y:2022:i:c:s154461232100146x

DOI: 10.1016/j.frl.2021.102065

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().