Optimal empirical Bayes estimation for the Poisson model via minimum-distance methods
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2022
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909784443191296 |
|---|---|
| author | Jana, Soham Polyanskiy, Yury Wu, Yihong |
| author_facet | Jana, Soham Polyanskiy, Yury Wu, Yihong |
| contents | The Robbins estimator is the most iconic and widely used procedure in the empirical Bayes literature for the Poisson model. On one hand, this method has been recently shown to be minimax optimal in terms of the regret (excess risk over the Bayesian oracle that knows the true prior) for various nonparametric classes of priors. On the other hand, it has been long recognized in practice that the Robbins estimator lacks the desired smoothness and monotonicity of Bayes estimators and can be easily derailed by those data points that were rarely observed before. Based on the minimum-distance distance method, we propose a suite of empirical Bayes estimators, including the classical nonparametric maximum likelihood, that outperform the Robbins method in a variety of synthetic and real data sets and retain its optimality in terms of minimax regret. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2209_01328 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Optimal empirical Bayes estimation for the Poisson model via minimum-distance methods Jana, Soham Polyanskiy, Yury Wu, Yihong Statistics Theory Methodology 62G05, 62G07 The Robbins estimator is the most iconic and widely used procedure in the empirical Bayes literature for the Poisson model. On one hand, this method has been recently shown to be minimax optimal in terms of the regret (excess risk over the Bayesian oracle that knows the true prior) for various nonparametric classes of priors. On the other hand, it has been long recognized in practice that the Robbins estimator lacks the desired smoothness and monotonicity of Bayes estimators and can be easily derailed by those data points that were rarely observed before. Based on the minimum-distance distance method, we propose a suite of empirical Bayes estimators, including the classical nonparametric maximum likelihood, that outperform the Robbins method in a variety of synthetic and real data sets and retain its optimality in terms of minimax regret. |
| title | Optimal empirical Bayes estimation for the Poisson model via minimum-distance methods |
| topic | Statistics Theory Methodology 62G05, 62G07 |
| url | https://arxiv.org/abs/2209.01328 |