Finding a Valid FX Covariance Matrix in the BS World
European University at St. Petersburg Department of Economics Working Paper No. EC-03/12
28 Pages Posted: 16 Dec 2012 Last revised: 6 Jan 2013
Date Written: June 1, 2012
Abstract
A number of methods has already been proposed for creating a valid correlation matrix in finance. However, such methods do not normally take into account additional restrictions on matrix elements imposed by specific non-arbitrage conditions in some markets, e.g. foreign exchange (FX). I suggest that taking those restrictions, known as triangular relationships, into account can lead to a more efficient method of correction of invalid correlation matrices, at least in FX markets. This paper outlines the steps of the new method.
Keywords: correlation matrix, eigenvalue, foreign exchange, triangular relationship, quantitative finance
JEL Classification: C63
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
Efficient Rank Reduction of Correlation Matrices
By Igor Grubisic and Raoul Pietersz
-
Efficient Rank Reduction of Correlation Matrices
By Igor Grubisic and Raoul Pietersz
-
Optimal Solution of the Nearest Correlation Matrix Problem by Minimization of the Maximum Norm
-
A Comparison of Single-Factor Markov-Functional and Multi-Factor Market Models
By Raoul Pietersz and Antoon Pelsser
-
A Comparison of Single Factor Markov-Functional and Multi Factor Market Models
By R. Pietersz and Antoon Pelsser
-
A Note on Solution of the Nearest Correlation Matrix Problem by von Neumann Matrix Divergence
-
Importance Sampling and Mm-Algorithms with Applications to Options Pricing