|
Activity Number:
|
333
- Topics in Reliability, Data Visualization and Modeling
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Quality and Productivity Section
|
|
Abstract #313720
|
|
|
Title:
|
Robust Estimation of Mean and Other Measures of Central Tendency Based on Weighted Samples
|
|
Author(s):
|
Emmanuel Yashchin*
|
|
Companies:
|
IBM Corporation
|
|
Keywords:
|
Estimation;
Trimmed mean;
Target setting;
Process capability ;
Contaminated data;
Resampling methods
|
|
Abstract:
|
We consider the problem of robust estimation of measures of central tendency when the data points have weights assigned to them. The problem was motivated by the target-setting task in early warning systems that focuses on evaluation of the process or vendor capability based on partially contaminated data with variable sample sizes. We introduce a methodology that generalizes the popular trimmed mean approach and compare it with existing methods for handling data of this type. We discuss several applications, including robust estimation for Binomial population with variable sample sizes and a very small parameter p.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2020 program
|
