International Finance Discussion Papers

Pitfalls in Tests for Changes in Correlations Brian H. Boyer, Michael S. Gibson, and Mico Loretan 1997-597

We show that such worries may not be justified since "correlation breakdowns" can easily be generated by data whose distribution is stationary and, in particular, whose correlation coefficient is constant. We make this point analytically, by way of several numerical examples, and via an empirical illustration.

But, risk managers should not necessarily relax. Although "correlation breakdown" can be an artifact of poor data analysis, other evidence suggests that correlations do in fact change over time, though not in a way that is correlated with "stressful" market conditions.

Full paper (695 KB PDF)

Keywords
Risk management, risk measurement, correlation, conditional correlation, normal distribution, foreign exchange, derivatives

PDF files: Adobe Acrobat Reader   ZIP files: PKWARE