|
Activity Number:
|
206
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #306464 |
|
Title:
|
Probabilistic Model To Evaluate Biological Process
|
|
Author(s):
|
Hrishikesh Chakraborty*+ and Pranab K. Sen
|
|
Companies:
|
RTI International and The University of North Carolina at Chapel Hill
|
|
Address:
|
Statistics Research Division, RTP, NC, 27709,
|
|
Keywords:
|
probabilistic model ; HIV transmission ; male-to-female ; bootstrap ; successive approximation ; underlying biological process
|
|
Abstract:
|
To model biological relationship it is impossible to test theoretical model due to availability of data. Relevant partial information can be obtained from different independent trials or from some published results. In this paper we demonstrated how to use conditional and unconditional probability theories to combine independent study data and published results to develop models to explain biological relationship. We used bootstrap re-sampling to account for repeated observations and successive approximation to estimate model parameters. This technique was used to develop two probabilistic models to evaluate the male-to-female penile-vaginal per-sexual-act HIV-1 transmission probability as a function of changes in transmitter cell and receptor cell counts. This type of modeling is very important when it is impractical to design one such study to evaluate underlying biological process.
|
