A tutorial will take place in a hybrid mode on Monday the 31st of July 2023. Participation is restricted only to those who attend the conference. The tutorial in-person venue is Room 201, Building 57, Nishi-Waseda Campus, 3-chome-4 Okubo, Shinjuku City, Tokyo 169-0072.

To access the tutorial online, click on the title below. Read the technical requirements and the general information that you will find in the virtual guidelines section carefully before entering the virtual room. Accessing this conference implies accepting the conditions. The password will be shown in the registration tool only to registered participants the day of the tutorial. Attendance is restricted to participants registered for the tutorial only. When joining on Zoom, please ensure to use your name as registered for the conference. Non-registered users will be promptly removed by the conference staff.

Confidence distribution: A new statistical inference approach and its applications in meta-analysis and fusion learning

Prof. Regina Liu, Rutgers, USA.
Email: Contact

Description:

The aim is to introduce the concept of confidence distribution, describe its role in statistical foundation in linking many of the existing inferential approaches, and show how it can be applied broadly to solve a wide range of problems in fusion learning and meta-analysis. Confidence distribution (CD) is a sample-dependent distribution function on the parameter space that can represent confidence intervals (regions) of all levels for a parameter of interest. It provides “simple and interpretable summaries of what can reasonably be learned from data (and an assumed model)” (Cox, 2013), and, in turn, meaningful answers to all questions related to statistical inference. In this short course, we review the development of CD and introduce the new and powerful inference tools derived from CD. The main focus is on combining information from different sources using confidence distributions. Specifically, we present several new and effective meta-analysis and fusion learning approaches, including 1) A unified framework for meta-analysis and software package gmeta; 2) Meta-analysis of heterogeneous studies using only summary statistics; 3) Incorporating external information in analyses of clinical trials with binary outcomes; 4) Exact meta-analysis approach for discrete data and its application to 2x2 tables with rare events; 5) Robust meta-analysis under the framework of CD combination; 6) Efficient network meta-analysis; 7) Meta-analysis with no model assumptions, 8) Nonparametric combing inferences from multiple sources. Many of these are motivated by real applications with real data, which are demonstrated throughout the course. Altogether, they show that CD can yield useful statistical inference tools for many statistical problems where methods with desirable properties have been lacking or are not easily available.