Abstract

We study the two-sample test problem on marginal tail features in

cluding the extreme value indices and scedasis functions in the presence of nonstationary tail dependence structures. To address this problem, we introduce a

unified bootstrap-based framework for bivariate heteroscedastic extremes, where

both margins and tail dependence structures are allowed to evolve over time. Our

approach is built upon a bivariate sequential tail empirical process, whose weak

convergence and bootstrap counterpart are established. Our simulations validate

the robustness and efficiency of the bootstrap-based tests. An empirical analysis is conducted on 8 assets, where two different scedasis functions are identified.

Information

Preprint No.SS-2025-0252
Manuscript IDSS-2025-0252
Complete AuthorsYifan Hu, Yanxi Hou
Corresponding AuthorsYanxi Hou
Emailsyxhou@fudan.edu.cn

References

  1. Adrian, T., N. Boyarchenko, and D. Giannone (2019). Vulnerable growth. American Economic Review 109(4), 1263–1289.
  2. Aloui, R., S. Hammoudeh, and D. K. Nguyen (2013). A time-varying copula approach to oil and stock market dependence: The case of transition economies. Energy economics 39, 208–221.
  3. Ando, T. and J. Bai (2020). Quantile co-movement in financial markets: A panel quantile model with unobserved heterogeneity. Journal of the American Statistical Association 115(529), 266–279.
  4. Ando, T., M. Greenwood-Nimmo, and Y. Shin (2022). Quantile connectedness: modeling tail behavior in the topology of financial networks. Management Science 68(4), 2401–2431.
  5. Bickel, P. J. and A. Sakov (2008). On the choice of m in the m out of n bootstrap and confidence bounds for extrema. Statistica Sinica, 967–985.
  6. Bücher, A. and H. Dette (2013). Multiplier bootstrap of tail copulas with applications. Bernoulli 19(5A), 1655–1687.
  7. Bücher, A. and T. Jennessen (2024). Statistics for heteroscedastic time series extremes. Bernoulli 30(1), 46–71.
  8. Bücher, A. and J. Segers (2018). Inference for heavy tailed stationary time series based on sliding blocks. Electronic Journal of Statistics 12(1), 1098–1125.
  9. Bücher, A. and I. Kojadinovic (2019). A Note on Conditional Versus Joint Unconditional Weak Convergence in Bootstrap Consistency Results. Journal of Theoretical Probability 32(3), 1145–1165.
  10. Chen, L., J. J. Dolado, and J. Gonzalo (2021). Quantile factor models. Econometrica 89(2), 875–910.
  11. Chernozhukov, V. and I. Fernández-Val (2011). Inference for extremal conditional quantile models, with an application to market and birthweight risks. The Review of Economic Studies 78(2), 559–589.
  12. de Haan, L. and C. Zhou (2024). Bootstrapping extreme value estimators. Journal of the American Statistical Association 119(545), 382–393.
  13. Drees, H. (2023). Statistical inference on a changing extreme value dependence structure. The Annals of Statistics 51(4), 1824–1849.
  14. Einmahl, J. H. and Y. He (2023). Extreme value inference for heterogeneous power law data. The Annals of Statistics 51(3), 1331–1356.
  15. Einmahl, J. H. and C. Zhou (2024). Tail copula estimation for heteroscedastic extremes. Econometrics and Statistics.
  16. Einmahl, J. H. J., L. Haan, and C. Zhou (2014). Statistics of Heteroscedastic Extremes. Journal of the Royal Statistical Society Series B: Statistical Methodology 78(1), 31–51.
  17. Engle, R. (2002). Dynamic conditional correlation: A simple class of multivariate generalized autoregressive conditional heteroskedasticity models. Journal of business & economic statistics 20(3), 339–350.
  18. Erb, C. B., C. R. Harvey, and T. E. Viskanta (1994). Forecasting international equity correlations. Financial analysts journal 50(6), 32–45.
  19. He, Y. and J. H. J. Einmahl (2024). Extreme value inference for general heterogeneous data. SSRN Electronic Journal.
  20. Jentsch, C. and R. Kulik (2021). Bootstrapping Hill estimator and tail array sums for regularly varying time series. Bernoulli 27(2), 1409–1439.
  21. Kinsvater, P., R. Fried, and J. Lilienthal (2016). Regional extreme value index estimation and a test of tail homogeneity. Environmetrics 27(2), 103–115.
  22. Kosorok, M. R. (2003). Bootstraps of sums of independent but not identically distributed stochastic processes. Journal of Multivariate Analysis 84(2), 299–318.
  23. Li, D., L. Peng, and Y. Qi (2011). Empirical likelihood confidence intervals for the endpoint of a distribution function. Test 20(2), 353–366.
  24. Longin, F. and B. Solnik (1995). Is the correlation in international equity returns constant: 1960–1990? Journal of international money and finance 14(1), 3–26.
  25. Mefleh, A., R. Biard, C. Dombry, and Z. Khraibani (2020). Trend detection for heteroscedastic extremes. Extremes 23(1), 85–115.
  26. Sarhadi, A., D. H. Burn, M. Concepcion Ausin, and M. P. Wiper (2016). Time-varying nonstationary multivariate risk analysis using a dynamic bayesian copula. Water Resources Research 52(3), 2327–2349.
  27. van der Vaart, A. W. and J. A. Wellner (1996). Weak Convergence and Empirical Processes: With Applications to Statistics. New York, NY: Springer.
  28. Xu, P., Z. Zhang, D. Wang, V. P. Singh, C. Zhang, X. Fu, and L. Wang (2023). A time-varying copula-based approach to quantify the effects of antecedent drought on hot extremes. Journal of Hydrology 627, 130418.
  29. Zhang, Y. (2018). Extremal quantile treatment effects. Annals of Statistics 46(6B), 3707–3740.
  30. Zou, N., S. Volgushev, and A. Bücher (2021). Multiple block sizes and overlapping blocks for multivariate time series extremes. Annals of Statistics 49(1), 295–320.

Acknowledgments

The authors gratefully acknowledge that the work is supported by the National Natural Science Foundation of China Grants 72171055.

Supplementary Materials

This supplement collects the proofs of the theoretical results in the article.

Supplementary materials are available for download.