Estimation of Effects of Sequential Treatments by Reparameterizing Directed Acyclic Graphs
arXiv:1302.1566
Abstract
The standard way to parameterize the distributions represented by a directed acyclic graph is to insert a parametric family for the conditional distribution of each random variable given its parents. We show that when one's goal is to test for or estimate an effect of a sequentially applied treatment, this natural parameterization has serious deficiencies. By reparameterizing the graph using structural nested models, these deficiencies can be avoided.
Appears in Proceedings of the Thirteenth Conference on Uncertainty in Artificial Intelligence (UAI1997)
References in corpus (1)
Cited by in corpus (15)
- Structural Nested Models and G-estimation: The Partially Realized Promise
- gfoRmula: An R package for estimating effects of general time-varying treatment interventions via the parametric g-formula
- Identifying the consequences of dynamic treatment strategies: A decision-theoretic overview
- Nested Markov Properties for Acyclic Directed Mixed Graphs
- Statistical modeling of causal effects in continuous time
- Causal Etiology of the Research of James M. Robins
- Revisiting the g-null paradox
- Polynomial Constraints in Causal Bayesian Networks
- Inequality Constraints in Causal Models with Hidden Variables
- The role of exchangeability in causal inference
- Causal Graphs: Addressing the Confounding Problem Without Instruments or Ignorability
- Causality from the Point of View of Statistics
- Causal Inference in the Time of Covid-19
- A causal approach to analysis of censored medical costs in the presence of time-varying treatment
- A Kernel Test for Causal Association via Noise Contrastive Backdoor Adjustment