Marginal Specification and Tail Dependence in Equity Markets
Main Article Content
Keywords
copula models, cross-market dependence, tail dependence, GARCH filtering, temporal alignment
Abstract
Copula conclusions for equity markets depend on both marginal filtering and the temporal pairing of daily returns. We analyse S&P 500, FTSE 100, and Nikkei 225 returns from 2010 to 2025 in three pairwise panels. A rolling 756-day training and 63-day evaluation design compares empirical-CDF, Gaussian-GARCH, and Student-t-GARCH margins. The GARCH filters use a standard one-step-ahead recursion, and every candidate copula is estimated by pseudo maximum likelihood. The Student-t copula has the highest window-weighted density score under all three margins, with scores of 0.0924, 0.0703, and 0.0685. It is the numerical winner in eight of nine panel–margin cells, but paired moving-block intervals support a positive advantage in only three cells. A lead–lag check materially changes the S&P 500–Nikkei 225 association, while a joint-downside probability score does not select the same family as the density score. The evidence therefore supports specification-aware conditional-dependence analysis, rather than an invariant model ranking or causal contagion claim.
References
- [1]Demarta, S., and McNeil, A. J., “The copula and related copulas,” International Statistical Review, 73, (1), 111–129, 2005, https://doi.org/10.1111/j.1751-5823.2005.tb00254.x.
- [2]Bollerslev, T., “Generalized autoregressive conditional heteroskedasticity,” Journal of Econometrics, 31, (3), 307–327, 1986, https://doi.org/10.1016/0304-4076(86)90063-1.
- [3]Münnix, M. C., and Schäfer, R., A copula approach on the dynamics of statistical dependencies in the US stock market, 2011, https://arxiv.org/abs/1102.1099.
- [4]Chen, Q., Wang, D., and Pan, M.-S., Multivariate time-varying copula GARCH model and its application in the financial market risk measurement, 2015, https://doi.org/10.1155/2015/286014.
- [5]Kim, Y., Hwang, S., Kim, J.-M., and Kim, S., Linear time-varying regression with copula–DCC–asymmetric–GARCH models for volatility: The co-movement between industrial electricity demand and financial factors, 2022, https://doi.org/10.1080/00036846.2022.2086684.
- [6]Das, B., and Fasen-Hartmann, V., On heavy-tailed risks under gaussian copula: The effects of marginal transformation, 2023, https://arxiv.org/abs/2304.05004.
- [7]Oh, M., and Kim, D., Effect of the u.s.–china trade war on stock markets: A financial contagion perspective, 2021, https://arxiv.org/abs/2111.09655.
- [8]Rigana, K., Wit, E. C., and Cook, S., Navigating market turbulence: Insights from causal network contagion value at risk, 2024, https://arxiv.org/abs/2402.06032.
- [9]Vasiljeva, A., Matvejevs, A., and Fjodorovs, J., Dependence modeling and portfolio optimization with copula-GARCH: A european investment perspective, 2025, https://doi.org/10.3389/fams.2025.1675120.
- [10]Cambriles, A. P., and Muela, S. B., Assessing the structure dependence between the spanish stock market and some international financial markets. A time-varying copula analysis., 2023, https://doi.org/10.32826/reyf.v1i1.342.
- [11]Tian, M., Khoury, R. E., and Alshater, M. M., The nonlinear and negative tail dependence and risk spillovers between foreign exchange and stock markets in emerging economies, 2022, https://doi.org/10.1016/j.intfin.2022.101712.
- [12]Mensi, W., Hanif, W., Bouri, E., and Vo, X., Spillovers and tail dependence between oil and US sectoral stock markets before and during COVID-19 pandemic, 2023, https://doi.org/10.1108/ijoem-12-2021-1799.
- [13]Cao, S., and Sun, Z., Extreme risk spillovers between chinese crude oil and stock markets: A CoVaR-copula approach, 2025, https://doi.org/10.23977/ferm.2025.080107.
- [14]Zhou, X., and Liu, H., Cross-market risk spillovers and tail dependence between u.s. And chinese technology-related equity markets, 2025, https://doi.org/10.3390/ijfs13040242.
- [15]Patton, A. J., “Modelling asymmetric exchange rate dependence,” International Economic Review, 47, (2), 527–556, 2006, https://doi.org/10.1111/j.1468-2354.2006.00387.x.
- [16]Gong, X.-L., Liu, X.-H., and Xiong, X., Measuring tail risk with GAS time varying copula, fat tailed GARCH model and hedging for crude oil futures, 2019, https://doi.org/10.1016/J.PACFIN.2019.03.010.
- [17]Tsai, J.-J., Wang, Y.-C., and Zhang, J., Measuring financial contagion between major european stock markets via the regime-switching vine copula model, 2018, https://doi.org/10.3934/dcdss.2020267.
- [18]Chen, W., Wei, Y., Lang, Q., Lin, Y., and Liu, M., Financial market volatility and contagion effect: A copula–multifractal volatility approach, 2014, https://doi.org/10.1016/J.PHYSA.2013.12.016.
- [19]Hansen, P. R., Lunde, A., and Nason, J. M., “The model confidence set,” Econometrica, 79, (2), 453–497, 2011, https://doi.org/10.3982/ECTA5771.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.