|
Activity Number:
|
399
- Modern Methods and Applications for HIV/AIDS Research
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #329978
|
Presentation
|
|
Title:
|
Statistical Considerations When Working with Mathematical Models
|
|
Author(s):
|
Sarah Holte and Chloe Krakauer*
|
|
Companies:
|
Fred Hutchinson Cancer Research Center and Universiry of Washington
|
|
Keywords:
|
mathematical models; parameter identifiability; model comparison; model validation; differential equations
|
|
Abstract:
|
Mathematical modeling using differential equations has become more and more common in studies of HIV -- both for transmission dynamics (population level) and viral and immune system dynamics (individual or aggregate level). However, appropriate statistical methods are not always used, and model validation is rarely performed, leading to potentially incorrect predictions by the proposed models This presentation will describe a number of statistical considerations which are appropriate for use when working with mathematical models. Specifically methods to determine parameter identifiability and comparison will be discussed and options for model validation will be considered.
|
Authors who are presenting talks have a * after their name.
