Fast inference methods for high-dimensional factor copulas
Gaussian factor models allow the statistician to capture multivariate dependence between variables. However, they are computationally cumbersome in high dimensions and are not able to capture multivariate skewness in the data. We propose a copula model that allows for arbitrary margins, and multivar...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2022-09-01
|
| Series: | Dependence Modeling |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/demo-2022-0117 |
| _version_ | 1797998408358690816 |
|---|---|
| author | Verhoijsen Alex Krupskiy Pavel |
| author_facet | Verhoijsen Alex Krupskiy Pavel |
| author_sort | Verhoijsen Alex |
| collection | DOAJ |
| description | Gaussian factor models allow the statistician to capture multivariate dependence between variables. However, they are computationally cumbersome in high dimensions and are not able to capture multivariate skewness in the data. We propose a copula model that allows for arbitrary margins, and multivariate skewness in the data by including a non-Gaussian factor whose dependence structure is the result of a one-factor copula model. Estimation is carried out using a two-step procedure: margins are modelled separately and transformed to the normal scale, after which the dependence structure is estimated. We develop an estimation procedure that allows for fast estimation of the model parameters in a high-dimensional setting. We first prove the theoretical results of the model with up to three Gaussian factors. Then, simulation results confirm that the model works as the sample size and dimensionality grow larger. Finally, we apply the model to a selection of stocks of the S&P500, which demonstrates that our model is able to capture cross-sectional skewness in the stock market data. |
| first_indexed | 2024-04-11T10:48:19Z |
| format | Article |
| id | doaj.art-f513ebf90c4249bb8cae822a15ba4003 |
| institution | Directory Open Access Journal |
| issn | 2300-2298 |
| language | English |
| last_indexed | 2024-04-11T10:48:19Z |
| publishDate | 2022-09-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Dependence Modeling |
| spelling | doaj.art-f513ebf90c4249bb8cae822a15ba40032022-12-22T04:28:59ZengDe GruyterDependence Modeling2300-22982022-09-0110127028910.1515/demo-2022-0117Fast inference methods for high-dimensional factor copulasVerhoijsen Alex0Krupskiy Pavel1School of Mathematics and Statistics, The University of Melbourne, Parkville, VIC 3010, AustraliaSchool of Mathematics and Statistics, The University of Melbourne, Parkville, VIC 3010, AustraliaGaussian factor models allow the statistician to capture multivariate dependence between variables. However, they are computationally cumbersome in high dimensions and are not able to capture multivariate skewness in the data. We propose a copula model that allows for arbitrary margins, and multivariate skewness in the data by including a non-Gaussian factor whose dependence structure is the result of a one-factor copula model. Estimation is carried out using a two-step procedure: margins are modelled separately and transformed to the normal scale, after which the dependence structure is estimated. We develop an estimation procedure that allows for fast estimation of the model parameters in a high-dimensional setting. We first prove the theoretical results of the model with up to three Gaussian factors. Then, simulation results confirm that the model works as the sample size and dimensionality grow larger. Finally, we apply the model to a selection of stocks of the S&P500, which demonstrates that our model is able to capture cross-sectional skewness in the stock market data.https://doi.org/10.1515/demo-2022-0117factor copulashigh-dimensional inferencefast inferencecomputational statisticsprimary 62h0562h12secondary 62-08 |
| spellingShingle | Verhoijsen Alex Krupskiy Pavel Fast inference methods for high-dimensional factor copulas Dependence Modeling factor copulas high-dimensional inference fast inference computational statistics primary 62h05 62h12 secondary 62-08 |
| title | Fast inference methods for high-dimensional factor copulas |
| title_full | Fast inference methods for high-dimensional factor copulas |
| title_fullStr | Fast inference methods for high-dimensional factor copulas |
| title_full_unstemmed | Fast inference methods for high-dimensional factor copulas |
| title_short | Fast inference methods for high-dimensional factor copulas |
| title_sort | fast inference methods for high dimensional factor copulas |
| topic | factor copulas high-dimensional inference fast inference computational statistics primary 62h05 62h12 secondary 62-08 |
| url | https://doi.org/10.1515/demo-2022-0117 |
| work_keys_str_mv | AT verhoijsenalex fastinferencemethodsforhighdimensionalfactorcopulas AT krupskiypavel fastinferencemethodsforhighdimensionalfactorcopulas |