|
Activity Number:
|
35
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #305616 |
|
Title:
|
Signed Directed Acyclic Graphs for Causal Inference
|
|
Author(s):
|
Tyler J. VanderWeele*+ and James Robins
|
|
Companies:
|
Harvard School of Public Health and Harvard School of Public Health
|
|
Address:
|
63 Mt. Vernon Street, 6, Cambridge, MA, 02140,
|
|
Keywords:
|
bias ; causal inference ; confounding ; directed acyclic graphs ; structural equations
|
|
Abstract:
|
By introducing the notions of a monotonic effect, a weak monotonic effect, and a signed edge, the directed acyclic graph causal framework can be extended to allow not only for the graphical representation of causal relations among variables, but also for the sign of these causal relations. Results are developed relating monotonic effects to the sign of the causal effect of an intervention in the presence of intermediate variables. Further, the incorporation of signed edges into the directed acyclic graph causal framework allows for the development of rules governing the relationship between monotonic effects and the sign of the covariance between two variables and rules governing the sign of the bias that arise when control for confounding is inadequate.
|
