UPDF AI

Causal Interpretation of Between-Within Models for Twin Research

A. Sjölander,T. Frisell,S. Öberg

2012 · DOI: 10.1515/2161-962X.1015
48 Citations

TLDR

It is shown that the BW model produces a specific subpopulation causal effect, in the absense of non-shared confounding, and that the causal interpretation can be retrieved with a minor modification of the model.