Online Program Home
  My Program Register!

Abstract Details

Activity Number: 83
Type: Invited
Date/Time: Sunday, July 30, 2017 : 8:30 PM to 10:30 PM
Sponsor: Section for Statistical Programmers and Analysts
Abstract #322649
Title: Approximate Confidence Distribution Computing: An Effective Likelihood-Free Method with Statistical Guarantees
Author(s): Suzanne Thornton* and Minge Xie
Companies: Rutgers University and Rutgers University
Keywords: approximate Bayesian computing ; confidence distribution ; Bernstein von Mises

Approximate Bayesian computing (ABC) is a likelihood-free method that has grown increasingly popular since early applications in population genetics. However, the theoretical justification for inference based on this method has yet to be fully developed especially pertaining to the use of non-sufficient summary statistics. We introduce a more general computational technique, approximate confidence distribution computing (ACC), to overcome a few issues associated with the ABC method, namely, the lack of theory supporting the use of non-sufficient summary statistics, the lack of guardian for the selection of prior, and the long computing time. We establish frequentist coverage properties for the outcome of ACC by using the theory of confidence distributions; thus inference based on ACC is justified even if one uses a non-sufficient summary statistic. Furthermore, the ACC method is very broadly applicable; in fact, the ABC algorithm can be viewed as a special case of ACC without damaging the integrity of ACC-based inference. We supplement the theory with simulation studies and an epidemiological application. We demonstrate the computational savings of a well-tended ACC algorithm.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association