The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
129
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2012 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
Abstract - #305307 |
Title:
|
Inference on Random Graphs with Classified Edge Attributes
|
Author(s):
|
Michael Trosset*+ and William David Brinda and Shantanu Jain
|
Companies:
|
Indiana University and Yale University and Indiana University
|
Address:
|
Statistics House, Bloomington, IN, 47408, United States
|
Keywords:
|
comparison of experiments ;
confusion matrices ;
stochastic matrices ;
partial ordering
|
Abstract:
|
We test simple hypotheses about a random graph with a fixed set of vertices and random edges, each of which possesses one of K mutually exclusive attributes. The edge attributes are inferred by means of a fallible classifier. Suppose that E and F are the confusion matrices of two such classifiers. Using results from statistical decision theory, we demonstrate that, if there exists a K x K stochastic matrix R such that ER = F, then most powerful (MP) tests based on E are necessarily more powerful than MP tests based on F. By means of an example, we also demonstrate that entry-wise superiority of E to F does not guarantee that an MP test based on E is more powerful than an MP test based on F.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.