Faculty Candidate Seminar: Efficient Empirical Bayes Variable Selection and Estimation

Event Details
  • Date/Time:
    • Friday February 20, 2004 - Thursday February 19, 2004
      9:00 am - 11:00 pm
  • Location: Executive Classroom (Main Bldg.)
  • Phone:
  • URL:
  • Email:
  • Fee(s):
    N/A
  • Extras:
Contact
Barbara Christopher
Industrial and Systems Engineering
Contact Barbara Christopher
404.385.3102
Summaries

Summary Sentence: Faculty Candidate Seminar: Efficient Empirical Bayes Variable Selection and Estimation

Full Summary: Faculty Candidate Seminar: Efficient Empirical Bayes Variable Selection and Estimation

Abstract: We propose an empirical Bayes method for variable selection and
coefficient estimation in linear regression models. The method is based on
a particular hierarchical Bayes formulation, and the estimator is shown to
be closely related to the LASSO estimator. Such a connection allows us to
take advantage of the recently developed quick LASSO algorithm to compute
the empirical Bayes estimate, and provides new ways to select the tuning
parameter in the LASSO method. Unlike previous empirical Bayes variable
selection methods, which in most practical situation can only be
implemented through a greedy stepwise algorithm, our method gives a global
solution efficiently. Simulations show that the proposed method compares
favorably with other variable selection and estimation methods in terms of
variable selection, estimation accuracy, and computation speed. This is
joint work with Professor Yi Lin.

Additional Information

In Campus Calendar
No
Groups

H. Milton Stewart School of Industrial and Systems Engineering (ISYE)

Invited Audience
No audiences were selected.
Categories
Seminar/Lecture/Colloquium
Keywords
No keywords were submitted.
Status
  • Created By: Barbara Christopher
  • Workflow Status: Published
  • Created On: Oct 8, 2010 - 7:42am
  • Last Updated: Oct 7, 2016 - 9:52pm