Abstract Details
Activity Number:
|
299
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Computing
|
Abstract #313565
|
View Presentation
|
Title:
|
Robust Regression by Self-Updating Process
|
Author(s):
|
Ting-Li Chen*+
|
Companies:
|
Academia Sinica
|
Keywords:
|
robust regression ;
weighted regression ;
iterative process
|
Abstract:
|
Robust regression aims to reduce the effect from outliers. One standard approach is to perform weighted regression in which the weights are iteratively updated according to the new fitted line. In this paper, we will present an iterative process to reduce the effect from outliers. It is an extension of SUP clustering algorithm (Chen and Shiu, 2007). This process updates both the weights and the data points through iterations. At each iteration, a line is fitted locally for each data point. The data point is then moved to this line. Throughout this process, all data points except outliers will gradually move to form a line. We will show results from simulation studies that our proposed method outperforms the standard approach. The success of the proposed algorithm comes from its two important properties: One is that the local estimation can reduce the effect from outliers so that the method is more robust. The other is that moving data based on the current estimation can improve the overall efficiency.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.