P Ding1, T J VanderWeele2, J M Robins2
1Department of Statistics, University of California, Berkeley, California, USA.
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Adjusting for more covariates in observational studies can increase bias when unmeasured confounding exists. This study provides a general theory for this "bias amplification" phenomenon beyond linear models.
Area of Science:
Background:
Purpose of the Study:
Main Methods:
Main Results:
Conclusions: