CMStatistics 2023: Start Registration
View Submission - CFE
A1404
Title: Regular vine copula-based portfolio optimization Authors:  Thomas Conlon - University College Dublin (Ireland)
Illia Kovalenko - University of Limerick (Ireland) [presenting]
John Cotter - University College Dublin (Ireland)
Abstract: The purpose is to study the use of the regular vine copula models in the context of the construction of portfolios with varying sizes. It is found that copula-based portfolios outperform the nave equally weighted benchmark before transaction costs. Significantly reduced tail risk makes copula-based portfolios especially desirable for investors with high-risk aversion. The superior performance is more pronounced during periods of high dependence asymmetry and market volatility. Sparse vine models, in which independence pair-copulas prevail, provide significantly improved results for large portfolios across various performance measures, specifically reducing turnover. However, the improvement of portfolio performance is attenuated as we consider transaction costs. Limiting large portfolio turnover by increasing investment horizon or rebalancing error tolerance restores the outperformance of copula-based strategies.