Estimation in the Presence of Noise of a Signal Which is Flat Except for Jumps. Part II. The Empirical Bayes Approach.
Abstract
This is the second of a two-part paper. In the first part Yao (1982), a special Bayesian Model A is studied in detail. In this part, a more general model is proposed and studied in an empirical Bayes framework. The results for Model A are applied to step-function signals using the ideas of empirical Bayes and maximum likelihood applied to the parameters of the Bayesian Model A. An efficient computational method is proposed to approximate the likelihood function under Model A. Several empirical Bayes estimators of the unknown step-function signal are compared by simulation. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1983
- Accession Number
- ADA124116
Entities
People
- Yi-ching Yao
Organizations
- Massachusetts Institute of Technology