MODELING INTRADAY PORTFOLIO RISK WITH DYNAMIC COPULAS AND ARTIFICIAL NEURAL NETWORKS

10.6292/AFPF.202212_(10).0005

Wing Ki Liu；Amanda M. Y. Chu；Mike K. P. So

Business Analytics ； Copula ； Artificial Neural Networks ； Time Varying Dependence ； Vine Decomposition

Advances in Financial Planning and Forecasting

10期（2022 / 12 / 01）

113 - 137

英文

The main purpose of this paper is to examine the application of dynamic copulas and artificial neural networks on multivariate intraday returns. The estimation of the parameters is a sequential process. In each of the experiment, we first apply artificial neural networks to estimate a marginal model and then apply dynamic copulas. In applying artificial neural networks, we found that periodicity is important in modeling the degree of freedom. In applying dynamic copulas, we examine whether the correlations are time-varying and what kind of dynamics the correlations exhibit. We found that the correlations for intraday returns are nonlinear and time-varying in two experiments. After applying dynamic copulas and artificial neural networks, we constructed portfolios based on the predicted means and covariance matrices. We found that the portfolios constructed outperform equally weighted portfolios in two experiments.

1. Abakah, E. J. A.,Addo, E., Jr.,Gil-Alana, L. A.,Tiwari, A. K.(2021).Re-examination of international bond market dependence: Evidence from a pair copula approach.International Review of Financial Analysis,74,101678.
2. Al Janabi, M. A. M.,Ferrer, R.,Shahzad, S. J. H.(2019).Liquidity-adjusted value-at-risk optimization of a multi-asset portfolio using a vine copula approach.Physica A: Statistical Mechanics and its Applications,536,122579.
3. Almeida, C.,Czado, C.,Manner, H.(2016).Modeling high-dimensional time-varying dependence using dynamic D-vine models.Applied Stochastic Models in Business and Industry,32,621-638.
4. Apergis, N.,Gozgor, G.,Lau, C. K. M.,Wang, S.(2020).Dependence structure in the Australian electricity markets: New evidence from regular vine copulae.Energy Economics,90,104834.
5. Bucci, A.(2020).Cholesky–ANN models for predicting multivariate realized volatility.Journal of Forecasting,39,865-876.
6. Çekin, S. E.,Pradhan, A. K.,Tiwari, A. K.,Gupta, R.(2020).Measuring co-dependencies of economic policy uncertainty in Latin American countries using vine copulas.The Quarterly Review of Economics and Finance,76,207-217.
7. Cherubini, U.,Gobbi, F.,Mulinacci, S.,Romagnoli, S.(2011).Dynamic copula methods in finance.New York, NY:Wiley.
8. Dai, X.,Wang, Q.,Zha, D.,Zhou, D.(2020).Multi-scale dependence structure and risk contagion between oil, gold, and US exchange rate: A wavelet-based vine-copula approach.Energy Economics,88,104774.
9. Dißmann, J.,Brechmann, E. C.,Czado, C.,Kurowicka, D.(2013).Selecting and estimating regular vine copulae and application to financial returns.Computational Statistics and Data Analysis,59,52-69.
10. Ji, H.,Wang, H.,Zhong, R.,Li, M.(2020).China’s liberalizing stock market, crude oil, and safe-haven assets: A linkage study based on a novel multivariate wavelet-vine copula approach.Economic Modelling,93,187-204.
11. Joe, H.(2015).Dependence modelling with copulas.Boca Raton, FL:CRC Press.
12. Karmakar, M.,Paul, S.(2019).Intraday portfolio risk management using VaR and CVaR: A CGARCH-EVT-Copula approach.International Journal of Forecasting,35,699-709.
13. Kim, H. Y.,Won, C. H.(2018).Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models.Expert Systems with Applications,103,25-37.
14. Kingma, D. P.,Ba, J. L.(2015).Adam: A method for stochastic optimization.3rd International Conference on Learning Representations (ICLR) 2015,San Diego, CA:
15. Kristjanpoller, W.,Minutolo, M. C.(2016).Forecasting volatility of oil price using an artificial neural network-GARCH model.Expert Systems with Applications,65,233-241.
16. Linton, O.,Wu, J.(2020).A coupled component DCS-EGARCH model for intraday and overnight volatility.Journal of Econometrics,217,176-201.
17. Liu, W. K.,So, M. K. P.(2020).A GARCH model with artificial neural networks.Information,11,1-17.
18. Liu, W. K.,So, M. K. P.,Chu, A. M. Y.(2021).Dynamic covariance modeling with artificial neural networks.Communications in Statistics: Case Studies, Data Analysis and Applications,8,15-42.
19. Naeem, M.,Umar, Z.,Ahmed, S.,Ferrouhi, E. M.(2020).Dynamic dependence between ETFs and crude oil prices by using EGARCH-Copula approach.Physica A: Statistical Mechanics and Its Applications,557,124885.
20. Sklar, M.(1959).Fonctions de répartition an dimensions et leurs marges.Publications de l’Institut de Statistique de Université de Paris,8,229-231.
21. So, M. K. P.,Yeung, C. Y. T.(2014).Vine-copula GARCH model with dynamic conditional dependence.Computational Statistics and Data Analysis,76,655-671.
22. Virbickaitė, A.,Ausín, M. C.,Galeano, P.(2020).Copula stochastic volatility in oil returns: Approximate Bayesian computation with volatility prediction.Energy Economics,92,104961.
23. Virbickaitė, A.,Frey, C.,Macedo, D. N.(2020).Bayesian sequential stock return prediction through copulas.Journal of Economic Asymmetries,22,00173.
24. Wang, W.,Li, W.,Zhang, N.,Liu, K.(2020).Portfolio formation with preselection using deep learning from long-term financial data.Expert Systems with Applications,143,113042.
25. Wen, D.,Wang, Y.,Zhang, Y.(2021).Intraday return predictability in China’s crude oil futures market: New evidence from a unique trading mechanism.Economic Modelling,96,209-219.
26. Xu, Y.,Bouri, E.,Saeed, T.,Wen, Z.(2020).Intraday return predictability: Evidence from commodity ETFs and their related volatility indices.Resources Policy,69,101830.
27. Yang, K.,Wei, Y.,Li, S.,He, J.(2020).Asymmetric risk spillovers between Shanghai and Hong Kong stock markets under China’s capital account liberalization.The North American Journal of Economics and Finance,51,101100.
28. Zhang, W.,Wang, P.,Li, Y.(2021).Bond intraday momentum.Journal of Behavioral and Experimental Finance,31,100515.
29. Zhu, B.,Zhou, X.,Liu, X.,Wang, H.,He, K.,Wang, P.(2020).Exploring the risk spillover effects among China’s pilot carbon markets: A regular vine copula-CoES approach.Journal of Cleaner Production,242,118455.