Causal inference from complex longitudinal data. (English) Zbl 0969.62072
Berkane, Maia (ed.), Latent variable modeling and applications to causality. Proceedings of the 1994 UCLA conference. New York, NY: Springer. Lect. Notes Stat. 120, 69-117 (1997).
In the eighties the author has introduced a formal theory of conterfactual causal inference [see, e.g., Math. Modelling 7, 1393-1512 (1986; Zbl 0614.62136)]. He has defined and studied causally interpreted structured tree graphs (CISTGs) and randomized CISTGs. Later (in 1995) he proved that the causal directed acyclic graphs (DAGs) introduced by Pearl, Spirtes, Glymour and Scheines are particular RCISTGs. This paper includes a summary and unification of the whole theory. Special attention is devoted to two methods of estimation: estimation of the conditional probabilities in the \(G\)-computation algorithm and \(G\)-estimation of structural nested models.
For the entire collection see [Zbl 0859.00026].
For the entire collection see [Zbl 0859.00026].
Reviewer: Urszula Wilczyńska (Łódź)
MSC:
| 62P10 | Applications of statistics to biology and medical sciences; meta analysis |
| 62P99 | Applications of statistics |
