Markov properties of nonrecursive causal models

By BY J. T. A. Koster. 1996

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Graphical models for marked point processes based on local independence (2007, unknown): “E-mail: vanessa.didelez@bristol.ac.uk systems (Spirtes, 1995; Koster, 1996) but are still based on cross-sectional data.”
Foundations of structural causal models with cycles and latent variables (2016, background): “The linear case is an extension of existing results for the linear-Gaussian setting without bidirected edges [28, 66, 67] to a linear (possibly non-Gaussian) setting with bidirected edges in the graph.”
Structural learning with time‐varying components: tracking the cross‐section of financial time series (2005, background): “g, h/ 1} ={h : h=g± e, for some e∈E}: The subset E.g/ of edges in E for which g±e is in D.g/ are called the legal edges. Here, E.g/≡E for all g; hence #D.g/=#E.g/≡#E =1+d.d −1/=2. However, when only some subclass G of the graphs is allowed, then E.g/ is usually a proper subset of E , defined in such”
Alternative Markov Properties for Chain Graphs (2001, background): “See Spirtes (1995) and Koster (1996) for related results.”
Structure learning in causal cyclic networks (2008, background): “Key developments (Spirtes, 1995) and (Koster, 1996), endowed DCGs with some of the properties of their DAG counterparts, and it was also shown that some SEMs are amenable to the same analysis (Spirtes, 1995; Pearl and Dechter, 1996).”
Joint Causal Inference from Multiple Contexts (2016, background): “The linear case extends existing results for the linearGaussian setting without latent comomn causes (Spirtes, 1994, 1995; Koster, 1996) to a linear (possibly non-Gaussian) setting with latent common causes.”