Wasserstein-based Minimax Estimation of Dependence in Multivariate Regularly Varying Extremes
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912112462266368 |
|---|---|
| author | Zhang, Xuhui Blanchet, Jose Marzouk, Youssef Nguyen, Viet Anh Wang, Sven |
| author_facet | Zhang, Xuhui Blanchet, Jose Marzouk, Youssef Nguyen, Viet Anh Wang, Sven |
| contents | We present the first minimax risk bounds for estimators of the spectral measure in multivariate linear factor models, where observations are linear combinations of regularly varying latent factors. Non-asymptotic convergence rates are derived for the multivariate Peak-over-Threshold estimator in terms of the $p$-th order Wasserstein distance, and information-theoretic lower bounds for the minimax risks are established. The convergence rate of the estimator is shown to be minimax optimal under a class of Pareto-type models analogous to the standard class used in the setting of one-dimensional observations known as the Hall-Welsh class. When the estimator is minimax inefficient, a novel two-step estimator is introduced and demonstrated to attain the minimax lower bound. Our analysis bridges the gaps in understanding trade-offs between estimation bias and variance in multivariate extreme value theory. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_09862 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Wasserstein-based Minimax Estimation of Dependence in Multivariate Regularly Varying Extremes Zhang, Xuhui Blanchet, Jose Marzouk, Youssef Nguyen, Viet Anh Wang, Sven Statistics Theory Methodology We present the first minimax risk bounds for estimators of the spectral measure in multivariate linear factor models, where observations are linear combinations of regularly varying latent factors. Non-asymptotic convergence rates are derived for the multivariate Peak-over-Threshold estimator in terms of the $p$-th order Wasserstein distance, and information-theoretic lower bounds for the minimax risks are established. The convergence rate of the estimator is shown to be minimax optimal under a class of Pareto-type models analogous to the standard class used in the setting of one-dimensional observations known as the Hall-Welsh class. When the estimator is minimax inefficient, a novel two-step estimator is introduced and demonstrated to attain the minimax lower bound. Our analysis bridges the gaps in understanding trade-offs between estimation bias and variance in multivariate extreme value theory. |
| title | Wasserstein-based Minimax Estimation of Dependence in Multivariate Regularly Varying Extremes |
| topic | Statistics Theory Methodology |
| url | https://arxiv.org/abs/2312.09862 |