VINE COPULA APPROACH FOR MULTIVARIATE AND MULTI-DAY AHEAD VALUE AT RISK AND EXPECTED SHORTFALL FORECASTING

Robina Iqbal ； Ghulam Sorwar ； Taufiq Choudhry

《Advances in Financial Planning and Forecasting》 10期 (2022/12) Pp. 163-196

https://doi.org/10.6292/AFPF.202212_(10).0007

摘要

In this paper, we construct vine copula models for multivariate stock portfolio returns to estimate one-day-ahead and multi-day ahead Value-at- Risk (VaR) and Expected Shortfall (ES) using Monte Carlo simulation. This is then compared with the VaR and ES using the dynamic conditional correlation (DCC) method. For the multi-day horizon, we use Monte Carlo simulation to simulate the share prices h-days ahead. The simulation-based method allows us to calculate VaR and ES for multivariate data at any horizons of interest and hence to calculate the entire term structure of risk. Using seven stocks from the DAX 30 as a case in point, we demonstrate the overall superiority of the copula-based method over the widely accepted DCC method. VaR and ES back-testing results indicate that vine copula significantly outperforms the DCC approach over a one-day horizon. This performance by the copula-based method is maintained across a multi-day horizon. Our findings suggest that institutions that use copula models to estimate their risk capital will need to set aside less capital to meet regulatory needs, than would otherwise be the case.

關鍵字

Finance ； Simulation ； Canonical Vine (C-vine) ； Drawable Vine (D-vine) ； Multi-Day Value-at-Risk (VaR) and Expected Shortfall (ES) ； Vine Copulas

參考文獻

Palaro, H., & Hotta, L. (2006). Using conditional copula to estimate value at risk. Journal of Data Science, 4, 93-115. doi:10.6339/JDS.2006.04(1).226

Gonzalez-Rivera, G., Lee, T. H., & Mishra, S. (2004). Forecasting volatility: A reality check based on option pricing, utility function, value-at-risk, and predictive likelihood. International Journal of Forecasting, 20, 629-645. doi:10.1016/j.ijforecast.2003.10.003

Haff, I. H. (2012). Comparison of estimators for pair-copula constructions. Journal of Multivariate Analysis, 110, 91-105. doi:10.1016/j.jmva.2011.08.013

Hartz, C., Mittnik, S., & Paolella, M. (2006). Accurate value-at-risk forecasting based on the normal-GARCH model. Computational Statistics and Data Analysis, 51, 2295-2312. doi:10.1016/j.csda.2006.09.017

Hoogerheide, L., & van Dijk, H. K. (2010). Bayesian forecasting of value at risk and expected shortfall using adaptive importance sampling. International Journal of Forecasting, 26, 231-247. doi:10.1016/j.ijforecast.2010.01.007