A general condition for avoiding effect reversal after marginalization

Journal of the Royal Statistical Society. Series B: Statistical Methodology

Volumn 65, Issue 4, 2003, Pages 937-941

  Cox, David R ^a   Wermuth, Nanny ^b  

^a NUFFIELD COLLEGE   (United Kingdom)

^b JOHANNES GUTENBERG UNIVERSITY   (Germany)

Author keywords

Causality; Collapsibility; Graphs; Randomization; Yule Simpson paradox

Indexed keywords

EID: 0242635644     PISSN: 13697412     EISSN: None     Source Type: Journal    
DOI: 10.1111/1467-9868.00424     Document Type: Article

References (9)

1. Cochran, W. G. (1938) The omission or addition of an independent variate in multiple linear regression. J. R. Statist. Soc., suppl., 5, 171-176.
2. Cox, D. R. and Wermuth, N. (1996) Multivariate Dependencies. London: Chapman and Hall.
3. Gail, M. H. (1988) The effect of pooling across strata in perfectly balanced studies. Biometrics, 44, 151-162.
4. Geng, Z. (1992) Collapsibility of relative risk in contingency tables with a response variable. J. R. Statist. Soc. B, 54, 585-593.
5. Lauritzen, S. L. (1996) Graphical Models. Oxford: Oxford University Press.
6. Rosenbaum, P. R. (2002) Observational Studies, 2nd edn. New York: Springer.
7. Snedecor, G. W. and Cochran, W. G. (1967) Statistical Methods. Ames: University of Iowa Press.
8. Wermuth, N. (1987) Parametric collapsibility and the lack of moderating effects in contingency tables with a dichotomous response variable. J. R. Statist. Soc. B, 49, 353-364.
9. Yule, G. U. (1903) Notes on the theory of association of attributes in statistics. Biometrika, 2, 121-134.