Growing semantic vines for robust asset allocation

Abstract

The vine structure has been widely studied as a graphical representation for high-dimensional dependence modeling, depicting complicated probability density functions, and robust correlation estimation. However, specification of the best vine structure is challenging as the number of candidate vine structures grows combinatorially when the number of elements increases. In this article, we propose to leverage semantic prior knowledge of assets extracted from their descriptive documents to find a suitable vine structure for financial portfolio optimization. A vine growing algorithm is provided and the robust covariance matrix estimation process is performed on this vine structure. Our construction of a semantic vine improves the state-of-the-art arbitrary vine-growing method in the context of robust correlation estimation and multi-period asset allocation. The effectiveness of our methods on a large scale is also demonstrated by experiments.