Introduction to Noise-Reduced Correlations Using Singular Spectrum Analysis

12 Pages Posted: 30 Aug 2017

See all articles by Jan Dash

Jan Dash

Fordham University; Bloomberg L.P.

Xipei Yang

Bloomberg L.P.

Mario Bondioli

Bloomberg L.P.

Harvey J. Stein

Two Sigma; Columbia University - Department of Mathematics

Date Written: August 28, 2017

Abstract

We summarize new results for estimating correlations for use in risk management. These estimates have better behavior than traditional estimation approaches from both a business standpoint and a technical standpoint. We smooth time series using Singular Spectrum Analysis (SSA) and compute correlations based on these smoothed series. We demonstrate that SSA-based correlation estimates have less noise than standard correlation estimates between unsmoothed series using: the signal-to-noise ratio, and distances from noise using polynomials generalizing the z-score and random matrix theory constructs. New useful analytic estimates for all eigenvalues of a random matrix are described. SSA-based correlations also enjoy superior time stability. Technical aspects are given in four accompanying papers, including extensive analyses of time stability and the noise-reduction tests described in this short paper.

Keywords: Singular Spectrum Analysis, Risk Management, Correlations, Stable, Noise-Cleaned, Polynomials Generalizing Z-Score, Signal-To-Noise Ratio, Random Matrix Theory, Analytic Eigenvalues of Random Matrix, Business Decisions

JEL Classification: C1, C14, C22, C63, E44, F65, G1, Y1

Suggested Citation

Dash, Jan and Yang, Xipei and Bondioli, Mario and Stein, Harvey J., Introduction to Noise-Reduced Correlations Using Singular Spectrum Analysis (August 28, 2017). Available at SSRN: https://ssrn.com/abstract=3028236 or http://dx.doi.org/10.2139/ssrn.3028236

Jan Dash (Contact Author)

Fordham University ( email )

113 West 60th Street
New York, NY 10023
United States

Bloomberg L.P. ( email )

731 Lexington Ave
New York, NY 10022
United States

Xipei Yang

Bloomberg L.P. ( email )

731 Lexington Avenue
New York, NY 10022
United States

Mario Bondioli

Bloomberg L.P. ( email )

731 Lexington Avenue
New York, NY 10022
United States

Harvey J. Stein

Two Sigma ( email )

100 6th Ave
New York, NY 10013
United States
10013 (Fax)

Columbia University - Department of Mathematics ( email )

New York, NY
United States

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

Paper statistics

Downloads
158
Abstract Views
880
Rank
338,002
PlumX Metrics