|
Activity Number:
|
324
- Causal Inference, Empirical Bayes and Related Topics in Regression
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #309806
|
|
|
Title:
|
Compound Empirical Bayes Interval Estimation
|
|
Author(s):
|
Wenhua Jiang* and Cun-Hui Zhang
|
|
Companies:
|
Fudan University and Rutgers University
|
|
Keywords:
|
Empirical Bayes;
interval estimation;
compound estimation;
generalized MLE;
Fourier transformation
|
|
Abstract:
|
We discuss empirical Bayes interval estimation under the compound framework. Two oracle rules are analyzed: the compound oracle rule and the Bayes oracle rule. The compound oracle rule is a better oracle in the sense of average interval length. We propose the smoothed generalized maximum likelihood estimator (GMLE) and the Fourier method to approximate the compound oracle rule. We prove the consistency of the smoothed GMLE interval and the convergence rate (\log n)^{-\alpha} for the Fourier interval when the unknown prior density is \alpha-smooth. Some numerical results are reported.
|
Authors who are presenting talks have a * after their name.
