|
Activity Number:
|
76
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 8:00 PM to 9:50 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #307129 |
|
Title:
|
Hierarchical Bayesian Calibration of Untested Devices
|
|
Author(s):
|
Reid Landes*+
|
|
Companies:
|
University of Arkansas for Medical Sciences
|
|
Address:
|
4301 W. Markham Street, # 781, Little Rock, AR, 72205,
|
|
Keywords:
|
measurement error ; MCMC ; prediction
|
|
Abstract:
|
We consider the problem of calibration of mass produced measuring devices, including the issue of extending inferences from some tested devices to a similar one for which no calibration data is available. Statistical methods are well-established for cases where the reference instrument is exact and inferences are needed only for tested devices. However, when the reference instrument is subject to measurement error, and particularly when inferences for an untested device are desired, new methods are needed. We study the properties of a method for producing such predictions of quantities from an untested device based on a Bayes hierarchical model applied in Landes (2005). We illustrate the method using a calibration experiment involving resistance temperature devices and an accurate, relatively precise thermometer. We evaluate the statistical properties of the method via simulation.
|
- 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 2006 program |
