|
Activity Number:
|
74
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #306882 |
|
Title:
|
Estimation of Transition Probabilities in a Discrete-Time Markov Chain with Missing Observations
|
|
Author(s):
|
Hung-Wen Yeh*+ and Wenyaw Chan
|
|
Companies:
|
The University of Texas School of Public Health and The University of Texas School of Public Health
|
|
Address:
|
1200 Herman Pressler, Houston, TX, 77030,
|
|
Keywords:
|
discrete-time Markov chain ; EM algorithm ; missing observation
|
|
Abstract:
|
The discrete-time Markov chain is commonly used in describing health states for chronic diseases. Statistical inferences on comparing treatment effects or on finding determinants of disease progression usually require estimation of transition probabilities. When collecting data of health states for these studies, researchers often encounter the problem of patients' occasional unavailability. Craig and Sendi (2002) use EM algorithm to estimate the transition probabilities when one scheduled observation was possibly missing in between two observed outcomes. In this research, EM algorithm will be applied to handle the situation when various numbers of scheduled observations are not observed between two measurements. An empirical study will be performed to examine the accuracy of the procedure and to compare the results with other methods. A real data set will be used for demonstration.
|
