Removing noise from correlations in multivariate stock price data
Statistical Mechanics
2008-12-02 v1 Statistical Finance
Abstract
This paper examines the applicability of Random Matrix Theory to portfolio management in finance. Starting from a group of normally distributed stochastic processes with given correlations we devise an algorithm for removing noise from the estimator of correlations constructed from measured time series. We then apply this algorithm to historical time series for the Standard and Poor's 500 index. We discuss to what extent the noise can be removed and whether the resulting underlying correlations are sufficiently accurate for portfolio management purposes.
Keywords
Cite
@article{arxiv.cond-mat/0403177,
title = {Removing noise from correlations in multivariate stock price data},
author = {Przemyslaw Repetowicz and Peter Richmond},
journal= {arXiv preprint arXiv:cond-mat/0403177},
year = {2008}
}
Comments
19 pages, 18 postscript figures, submitted to Physica A in March 2004