The Gerber Statistic: A Robust Measure of Correlation

7 Pages Posted: 6 May 2019

See all articles by Sander Gerber

Sander Gerber

Hudson Bay Capital Management, LP

Babak Javid

Hudson Bay Capital Management, LP

Harry Markowitz

University of California at San Diego

Paul Sargen

Hudson Bay Capital Management, LP

David Starer

Stevens Institute of Technology

Date Written: February 21, 2019

Abstract

We introduce the Gerber statistic, a robust measure of correlation. The statistic extends Kendall's Tau by counting the proportion of simultaneous co-movements in series when their amplitudes exceed data-dependent thresholds. This is unlike the standard Pearson correlation that is sensitive to outliers or the Spearman correlation that relies on ranking observations. Since the statistic is neither affected by extremely large or extremely small movements, it is especially suited to financial time series since these can exhibit extreme movements as well as a great amount of noise. Therefore, the statistic can advantageously be converted into a robust estimate of a covariance matrix that is suitable for portfolio optimization.

Suggested Citation

Gerber, Sander and Javid, Babak and Markowitz, Harry and Sargen, Paul and Starer, David, The Gerber Statistic: A Robust Measure of Correlation (February 21, 2019). Available at SSRN: https://ssrn.com/abstract=3351212 or http://dx.doi.org/10.2139/ssrn.3351212

Sander Gerber

Hudson Bay Capital Management, LP ( email )

777 Third Avenue
New York, NY NY 10017
United States

Babak Javid

Hudson Bay Capital Management, LP ( email )

777 Third Avenue
New York, NY NY 10017
United States

Harry Markowitz

University of California at San Diego ( email )

9500 Gilman Drive
La Jolla, CA 92093-0508
United States
(858) 534-3383 (Phone)

Paul Sargen

Hudson Bay Capital Management, LP ( email )

777 Third Avenue
New York, NY NY 10017
United States

David Starer (Contact Author)

Stevens Institute of Technology

Castle Point on Hudson
Hoboken, NJ 07030
United States

HOME PAGE: http://personal.stevens.edu/~dstarer

Register to save articles to
your library

Register

Paper statistics

Downloads
48
Abstract Views
265
PlumX Metrics
!

Under construction: SSRN citations will be offline until July when we will launch a brand new and improved citations service, check here for more details.

For more information