Motivated by the use of high-dimensional data such as data from several hundred risk-factor changes in the realm of quantitative risk management, we raise the following simple question, namely, How can one detect and visualize dependence in high-dimensional data?
Research seminar given at the Institute for Statistics and Mathematics, Wien University.
The structure of a set of high dimensional data objects (e.g. images, documents, molecules, genetic expressions, etc.) is notoriously difficult to visualize. In contrast, lower dimensional structures (esp. 3 or fewer dimensions) are natural to us and …