EFFECT OF MISSING AN INFLUENTIAL COVARIATE
- 30 April 2001
- journal article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 30 (5) , 837-853
- https://doi.org/10.1081/sta-100002261
Abstract
It is well known that while considering relationship between a response variate Yand an auxiliary variate X, if another important covariate Zis ignored, we may get into a paradoxical situation. One such example is Simpson's paradox. In this paper, we investigate the extent of effect of ignoring an important covariate Zon the relationship between the response variate Yand the auxiliary variate X. The relationship between these variates is modelled by logistic regression model and Cox regression model.Keywords
This publication has 12 references indexed in Scilit:
- Simpson-type paradoxes, dependence, and ageingJournal of Applied Probability, 1999
- A geometric approach to assess bias due to omitted covariates in generalized linear modelsBiometrika, 1993
- Simpson's Paradox and Related PhenomenaJournal of the American Statistical Association, 1993
- Homogeneity of Subpopulations and Simpson's ParadoxJournal of the American Statistical Association, 1991
- The Amalgamation and Geometry of Two-by-Two Contingency TablesThe Annals of Statistics, 1987
- Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariatesBiometrika, 1984
- Simpson's Paradox and Mutually Favorable EventsJournal of the American Statistical Association, 1973
- The Effect of 6-Mercaptopurine on the Duration of Steroid-induced Remissions in Acute Leukemia: A Model for Evaluation of Other Potentially Useful TherapyBlood, 1963
- Nonparametric Estimation from Incomplete ObservationsJournal of the American Statistical Association, 1958
- NOTES ON THE THEORY OF ASSOCIATION OF ATTRIBUTES IN STATISTICSBiometrika, 1903