Description Details Author(s) See Also
This package is for estimation of probability associated with collective counterfactual outcomes using approximation via causal graphical model. We apply a parsimonious parameterization for social network data with some specific kinds of interference and contagion, which corresponds to particular family of graphical models known as chain graphs.
We provide functions to estimate the parameters in conditional log-linear model when the observations (outcomes, treatments, and confounders) and the structure of a causal graph is given. Based on the estimated parameters, we generate counterfactual outcomes using Gibbs sampling to infer the causal effect (or causal probability) of a certain treatment assignment on the collective outcomes. Moreover, we use this method to identify causally influential units on social network.
Youjin Lee
Maintainer: Youjin Lee <ylee160@jhu.edu>
https://github.com/youjin1207/netchain
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.