Abstract Details
Activity Number:
|
580
|
Type:
|
Invited
|
Date/Time:
|
Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract #310561
|
View Presentation
|
Title:
|
Simplification of Agent-Based Epidemic Models
|
Author(s):
|
Georgiy Bobashev*+ and Daniel Heard and Robert J. Morris
|
Companies:
|
RTI International and Duke University and RTI International
|
Keywords:
|
Epidemic model ;
Stochastic process ;
Agent-based model ;
Model simplification ;
Model comparison ;
Statistical equivalence
|
Abstract:
|
Agent-based models (ABMs) allow one to directly model causal relationships in individual behavior and social interactions. While local behavior of the agents could be quite complex, model outcome is often defined at the aggregated (e.g. population) level. The challenge is to develop a model that is simpler and more interpretable but at the same time producing results equivalent to an output of more complex and higher fidelity model. We present an approach where complex structure of the underlying ABM (e.g. social networks, learning, non-linear interactions between multiple agents) could be collapsed to a set of simpler emulating models producing statistically equivalent trajectories. We discuss statistical challenges accompanying the development of the emulators and show that simple statistical emulators might be insufficient to describe adaptive behavior, especially when the model is supposed to describe response to interventions. We illustrate our method on an example of hybrid (agent-based/equation-based) models for infectious diseases and an example of adaptive emulators for drug use which combines the features of infectious and chronic diseases.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.