×

zbMATH — the first resource for mathematics

Are there algorithms that discover causal structure? (English) Zbl 1157.62310
Summary: There have been many efforts to infer causation from association by using statistical models. Algorithms for automating this process are a more recent innovation. In a former paper [Br. J. Philos. Sci. 47, 113–123 (1996)] we showed that one such approach, by Spirtes et al. [ibid. 48, 555–568 (1997)], was fatally flawed. Here we put our arguments in a broader context and reply to Korb and Wallace [ibid., 543–553 (1997)] and to Spirtes et al. Their arguments leave our position unchanged: claims to have developed a rigorous engine for inferring causation from association are premature at best, the theorems have no implications for samples of any realistic size, and the examples used to illustrate the algorithms are indicative of failure rather than success. The gap between association and causation has yet to be bridged.

MSC:
62A01 Foundations and philosophical topics in statistics
03A05 Philosophical and critical aspects of logic and foundations
65C60 Computational problems in statistics (MSC2010)
PDF BibTeX XML Cite
Full Text: DOI