Sample size importantly limits the usefulness of instrumental variable methods, depending on instrument strength and level of confounding

Author(s):
Boef, A.G.C.; Dekkers, O.M.; Vandenbroucke, J.P. (0000-0001-5668-6716); Cessie, S. le (0000-0003-2154-4923)
Published
2014
Publication type
Article / Letter to editor
Instrumental variableObservational studiesConfoundingVarianceSimulation studyTherapeutic effects
Date issued
2014-11-30
Language
English
Is part of
Journal of Clinical Epidemiology (2014), Vol. 67, no. 11, p. 1258-1264
hdl
1887/101817
doi
10.1016/j.jclinepi.2014.05.019