Mastering the Dimensions of Correlations
Hens Steehouwer, Ortec Finance

Correlations are an essential component of a risk management and investment decision framework. They determine how risk aggregates across different asset classes and liabilities into portfolio or balance sheet risk. Unfortunately, correlations are very complex because they have many dimensions, and calibrating risk models to realistic correlation structures can be difficult and time-consuming. Practitioners often follow a “partial approach” in which they calibrate models on an application-specific subset of the correlation dimensions. However, such a partial approach is inefficient and inconsistent. In this paper, we discuss the various dimensions of correlations and illustrate how to master these with the help of well-designed and calibrated risk models.