An Alternative Prior for Estimation in High-Dimensional Settings
Bibliographic Data
| ID | 21641860 |
|---|---|
| Authors | Michael Nagel (0000-0001-8436-8643, Eberhard Karls Universität), Lukáš Fischer (0000-0003-1852-0461, Eberhard Karls Universität), Tim Pawlowski (0000-0001-5829-963X, Eberhard Karls Universität), Augustin Kelava (0000-0001-6053-0415, Eberhard Karls Universität) |
| Year | 2024 |
| Volume | 31 |
| Issue | 6 |
| Pages | 939-951 |
| Publication date | 2024-11-01 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Structural Equation Modeling: A Multidisciplinary Journal (JOURNAL) |
| Journal identifiers | ISSN: 1070-5511 • E-ISSN: 1532-8007 |
| Publisher | Informa UK Limited (PUBLISHER • GB) |
| DOI | 10.1080/10705511.2023.2281279 |
| OpenAlex | W4400002290 |
| Language | EN |
| References cited | 29 |
Bayesian estimations of complex regression models with high-dimensional parameter spaces require advanced priors, capable of addressing both sparsity and multicollinearity in the data. The Dirichlet-horseshoe, a new prior distribution that combines and expands on the concepts of the regularized horseshoe and the Dirichlet-Laplace priors, is a novel approach that offers a high degree of flexibility and yields estimates with comparably high accuracy. To evaluate its performance in different frameworks, this study reports on two replicated simulation studies and a real-data example. Across all tested settings, the proposed approach outperforms or achieves similar performance to well-established regularization priors in terms of loss
Econometrics · Economics · Estimation · Advanced Causal Inference Techniques · Computer Science · Mathematics · Statistical Methods and Bayesian Inference · Statistical Methods and Inference
| Highly cited | No |
|---|