The Misleading Value of Measured Correlation

Wilmott, Volume 2012, Issue 62, pages 64-73, November 2012

10 Pages Posted: 15 Feb 2017

See all articles by Babak Mahdavi-Damghani

Babak Mahdavi-Damghani

University of Oxford - Oxford-Man Institute of Quantitative Finance

Daniella Welch Welch

None

Ciaran O'Malley

None

Stephen Knights

University of Oxford - Department of Economics

Date Written: 2012

Abstract

Within the framework of the financial industry, when representing relationships between assets, correlation is typically used. However, academics have long since questioned this method due to the plethora of issues that plague it. Indeed, it is thought that cointegration is a natural replacement in some of the cases as it is able to represent the physical reality of these assets better. However, despite this general academic consensus, financial practitioners refuse to accept cointegration as a better tool, or even the lesser of two evils. This technical report attempts to explain this bias, specifically focusing on the various consequences of model selection considering the new and challenging regulatory environment and suggests a practical replacement hybrid alternative to both cointegration and correlation.

Keywords: correlation; cointegration; mean reversion; cointelation

Suggested Citation

Mahdavi-Damghani, Babak and Welch, Daniella Welch and O'Malley, Ciaran and Knights, Stephen, The Misleading Value of Measured Correlation (2012). Wilmott, Volume 2012, Issue 62, pages 64-73, November 2012. Available at SSRN: https://ssrn.com/abstract=2429142

Babak Mahdavi-Damghani (Contact Author)

University of Oxford - Oxford-Man Institute of Quantitative Finance ( email )

United Kingdom

Stephen Knights

University of Oxford - Department of Economics ( email )

Manor Road Building
Manor Road
Oxford, OX1 3BJ
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
204
Abstract Views
884
rank
149,053
PlumX Metrics