Improving bias and coverage in instrumental variable analysis with weak instruments for continuous and binary outcomes
Abstract
Causal estimates can be obtained by instrumental variable analysis using a two‐stage method. However, these can be biased when the instruments are weak. We introduce a Bayesian method, which adjusts for the first‐stage residuals in the second‐stage regression and has much improved bias and coverage properties. In the continuous outcome case, this adjustment reduces median bias from weak instruments to close to zero. In the binary outcome case, bias from weak instruments is reduced and the estimand is changed from a marginal population‐based effect to a conditional effect. The lack of distributional assumptions on the posterior distribution of the causal effect gives a better summary of uncertainty and more accurate coverage levels than methods that rely on the asymptotic distribution of the causal estimate. We discuss these properties in the context of Mendelian randomization. Copyright © 2012 John Wiley & Sons, Ltd.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Cites
(2001) · RePEc: Research Papers in Economics · Cited by 27,577 · see citing works
Mostly Harmless Econometrics: An Empiricist's Companion
Angrist, J. D., & Pischke, J.-S. (2009) · doi:10.1515/9781400829828 · Cited by 9,467 · see citing works
Testing for Weak Instruments in Linear IV Regression
Stock, J. H., & Yogo, M. (2005) · Identification and Inference for Econometric Models · doi:10.1017/cbo9780511614491.006 · Cited by 7,211 · see citing works
Instrumental Variables Regression with Weak Instruments
Staiger, D., & Stock, J. H. (1997) · Econometrica · doi:10.2307/2171753 · Cited by 7,081 · see citing works
Davey Smith, G., & Ebrahim, S. (2003) · International Journal of Epidemiology · doi:10.1093/ije/dyg070 · Cited by 5,882 · see citing works
Estimation and Inference in Econometrics.
Chambers, M. J., Davidson, R., & MacKinnon, J. G. (1994) · The Economic Journal · doi:10.2307/2234656 · Cited by 5,536 · see citing works
Instrumental Variables Regression with Weak Instruments
Staiger, D., & Stock, J. (1994) · doi:10.3386/t0151 · Cited by 5,302 · cited 2× · see citing works
“from ordinary least squares regression of Y onX) [12], is asymptotically approximately 1/F , where F is the expected F statistic for the IVs in the first-stage regression [12, 15].”
BackgroundMendelian randomization: Using genes as instruments for making causal inferences in epidemiology
Lawlor, D. A., Harbord, R. M., Sterne, J. A. C., Timpson, N., & Davey Smith, G. (2008) · Statistics in Medicine · doi:10.1002/sim.3034 · Cited by 4,618 · see citing works
A Survey of Weak Instruments and Weak Identification in Generalized Method of Moments
Stock, J. H., Wright, J. H., & Yogo, M. (2002) · Journal of Business and Economic Statistics · doi:10.1198/073500102288618658 · Cited by 4,151 · see citing works
Bound, J., Jaeger, D. A., & Baker, R. (1995) · Journal of the American Statistical Association · doi:10.1080/01621459.1995.10476536 · Cited by 3,845 · see citing works
Instrumental Variables and GMM: Estimation and Testing
Baum, C. F., Schaffer, M. E., & Stillman, S. (2003) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0300300101 · Cited by 2,632 · see citing works
Robins, J. (1986) · Mathematical Modelling · doi:10.1016/0270-0255(86)90088-6 · Cited by 2,451 · see citing works
Testing for Weak Instruments in Linear IV Regression
Stock, J., & Yogo, M. (2002) · doi:10.3386/t0284 · Cited by 1,842 · see citing works
Two-stage residual inclusion estimation: Addressing endogeneity in health econometric modeling
Terza, J. V., Basu, A., & Rathouz, P. J. (2008) · Journal of Health Economics · doi:10.1016/j.jhealeco.2007.09.009 · Cited by 1,728 · see citing works
Testing Identifiability and Specification in Instrumental Variable Models
Cragg, J. G., & Donald, S. G. (1993) · Econometric Theory · doi:10.1017/s0266466600007519 · Cited by 1,464 · see citing works
Mendelian randomization: prospects, potentials, and limitations
Smith, G. D. (2004) · International Journal of Epidemiology · doi:10.1093/ije/dyh132 · Cited by 1,222 · see citing works
Some Problems in Interval Estimation
Fieller, E. C. (1954) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/j.2517-6161.1954.tb00159.x · Cited by 1,159 · see citing works
An introduction to instrumental variables for epidemiologists
Greenland, S. (2000) · International Journal of Epidemiology · doi:10.1093/ije/29.4.722 · Cited by 1,156 · see citing works
Bound, J., Jaeger, D. A., & Baker, R. M. (1995) · Journal of the American Statistical Association · doi:10.2307/2291055 · Cited by 1,018 · see citing works
Cuba—a library for multidimensional numerical integration
Hahn, T. (2005) · Computer Physics Communications · doi:10.1016/j.cpc.2005.01.010 · Cited by 962 · see citing works
Confounding and Collapsibility in Causal Inference
Greenland, S., Pearl, J., & Robins, J. M. (1999) · Statistical Science · doi:10.1214/ss/1009211805 · Cited by 933 · cited 2× · see citing works
“A measure of association is collapsible over a variable if it is constant across the strata of the variable, and if this constant value equals the value obtained from the marginal analyses [16].”
BackgroundThe Fitting of Straight Lines if Both Variables are Subject to Error
Wald, A. (1940) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177731868 · Cited by 930 · see citing works
Mendelian randomization as an instrumental variable approach to causal inference
Didelez, V., & Sheehan, N. (2007) · Statistical Methods in Medical Research · doi:10.1177/0962280206077743 · Cited by 923 · see citing works
C Reactive Protein Coronary Heart Disease Genetics Collaboration (CCGC) (2011) · BMJ · doi:10.1136/bmj.d548 · Cited by 642 · see citing works
INTERPRETATION AND CHOICE OF EFFECT MEASURES IN EPIDEMIOLOGIC ANALYSES1
Greenland, S. (1987) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a114593 · Cited by 546 · see citing works
Correcting for non-compliance in randomized trials using structural nested mean models
Robins, J. M. (1994) · Communication in Statistics- Theory and Methods · doi:10.1080/03610929408831393 · Cited by 534 · see citing works
TEACHER'S CORNER: Structural Equation Modeling With the sem Package in R
Fox, J. (2006) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1207/s15328007sem1303_7 · Cited by 433 · see citing works
Bias in causal estimates from Mendelian randomization studies with weak instruments
Burgess, S., & Thompson, S. G. (2011) · Statistics in Medicine · doi:10.1002/sim.4197 · Cited by 421 · see citing works
Self-citationNelson, C. R., & Startz, R. (1990) · The Journal of Business · doi:10.1086/296497 · Cited by 380 · see citing works
Nelson, C. R., & Startz, R. (1988) · National Bureau of Economic Research · doi:10.3386/t0069 · Cited by 364 · see citing works
Estimation with weak instruments: Accuracy of higher‐order bias and MSE approximations
Hahn, J., Hausman, J. A., & Kuersteiner, G. M. (2004) · Econometrics Journal · doi:10.1111/j.1368-423x.2004.00131.x · Cited by 342 · see citing works
An adaptive algorithm for the approximate calculation of multiple integrals
Berntsen, J., Espelid, T. O., & Genz, A. C. (1991) · ACM Transactions on Mathematical Software · doi:10.1145/210232.210233 · Cited by 314 · see citing works
A strategy for modelling the effect of a continuous covariate in medicine and epidemiology.
Royston, P. (2000) · Statistics in Medicine · doi:10.1002/1097-0258(20000730)19:14 · Cited by 234 · see citing works
Assumptions of IV Methods for Observational Epidemiology.
Didelez, V., Meng, S., & Sheehan, N. A. (2010) · Statistical Science · doi:10.1214/09-sts316 · Cited by 214 · see citing works
Robust, Accurate Confidence Intervals with a Weak Instrument: Quarter of Birth and Education
Imbens, G. W., & Rosenbaum, P. R. (2004) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/j.1467-985x.2004.00339.x · Cited by 171 · see citing works
Instrumental Variable Analysis for Estimation of Treatment Effects With Dichotomous Outcomes
Rassen, J. A., Schneeweiß, S., Glynn, R., Mittleman, M. A., & Brookhart, M. A. (2008) · American Journal of Epidemiology · doi:10.1093/aje/kwn299 · Cited by 164 · see citing works
Bayesian and classical approaches to instrumental variable regression
Kleibergen, F. R., & Zivot, E. W. (2003) · Journal of Econometrics · doi:10.1016/s0304-4076(02)00219-1 · Cited by 153 · see citing works
Instrumental Variable Estimators for Binary Outcomes
Clarke, P. S., & Windmeijer, F. A. G. (2012) · Journal of the American Statistical Association · doi:10.1080/01621459.2012.734171 · Cited by 149 · see citing works
On Instrumental Variables Estimation of Causal Odds Ratios
Vansteelandt, S., Bowden, J., Babanezhad, M., & Goetghebeur, E. J. T. (2011) · Statistical Science · doi:10.1214/11-sts360 · Cited by 116 · see citing works
Two‐stage instrumental variable methods for estimating the causal odds ratio: Analysis of bias
Cai, B., Small, D. S., & Have, T. R. T. (2011) · Statistics in Medicine · doi:10.1002/sim.4241 · Cited by 109 · see citing works
Valid Confidence Intervals and Inference in the Presence of Weak Instruments
Zivot, E. W., Startz, R., & Nelson, C. R. (1998) · International Economic Review · doi:10.2307/2527355 · Cited by 105 · see citing works
Instrumental Variables for Logistic Regression: An Illustration
Foster, E. M. (1997) · Social Science Research · doi:10.1006/ssre.1997.0606 · Cited by 104 · see citing works
Lawlor, D. A., Harbord, R. M., Timpson, N. J., Lowe, G. D. O., Rumley, A., Gaunt, T. R., Baker, I. A., YARNELL, J. W. G., Mika Kivimaki, B., Meena Kumari, B., et al. (2008) · PLoS ONE · doi:10.1371/journal.pone.0003011 · Cited by 104 · see citing works
The Existence of Moments of k-Class Estimators
Kinal, T. W. (1980) · Econometrica · doi:10.2307/1912027 · Cited by 99 · see citing works
Tests and Confidence Sets with Correct Size when Instruments are Potentially Weak
Mikusheva, A., & Poi, B. P. (2006) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0600600303 · Cited by 99 · see citing works
‘Mendelian randomization’ equals instrumental variable analysis with genetic instruments
Wehby, G. L., Ohsfeldt, R. L., & Murray, J. C. (2008) · Statistics in Medicine · doi:10.1002/sim.3255 · Cited by 90 · see citing works
Valid Confidence Intervals and Inference in the Presence of Weak InstrumentsPreprint
Nelson, C. R., Startz, R., & Zivot, E. W. (1996) · RePEc: Research Papers in Economics · Cited by 87 · see citing works
Palmer, T., Thompson, J. R., Tobin, M. D., Sheehan, N. A., & Burton, P. R. (2008) · International Journal of Epidemiology · doi:10.1093/ije/dyn080 · Cited by 87 · see citing works
Combining MCMC with ‘sequential’ PKPD modelling
Lunn, D. J., Best, N., Spiegelhalter, D. J., Graham, G., & Neuenschwander, B. E. (2009) · Journal of Pharmacokinetics and Pharmacodynamics · doi:10.1007/s10928-008-9109-1 · Cited by 86 · see citing works
Robust confidence sets in the presence of weak instruments
Mikusheva, A. (2010) · Journal of Econometrics · doi:10.1016/j.jeconom.2009.12.003 · Cited by 84 · see citing works
Minelli, C. (2004) · American Journal of Epidemiology · doi:10.1093/aje/kwh228 · Cited by 78 · see citing works
Johnston, K., Gustafson, P., Levy, A. R., & Grootendorst, P. V. (2007) · Statistics in Medicine · doi:10.1002/sim.3036 · Cited by 72 · see citing works
Zohoori, N., & Savitz, D. A. (1997) · Annals of Epidemiology · doi:10.1016/s1047-2797(97)00023-9 · Cited by 72 · see citing works
Tobin, M. D. (2004) · International Journal of Epidemiology · doi:10.1093/ije/dyh016 · Cited by 68 · see citing works
Estimation of Bias in Nongenetic Observational Studies Using “Mendelian Triangulation”
Bautista, L. E., Smeeth, L. L., Hingorani, A. D., & Casas, J. P. (2006) · Annals of Epidemiology · doi:10.1016/j.annepidem.2006.02.001 · Cited by 67 · see citing works
Greenland, S., Lanes, S., & Jara, M. (2008) · Clinical Trials · doi:10.1177/1740774507087703 · Cited by 62 · see citing works
Chapter 4. Instrumental Variables in Action: Sometimes You Get What You Need
Angrist, J. D., & Pischke, J. (2009) · Princeton University Press eBooks · doi:10.1515/9781400829828-009 · Cited by 58 · see citing works
Fischer-Lapp, K., & Goetghebeur, E. J. T. (1999) · Controlled Clinical Trials · doi:10.1016/s0197-2456(99)00027-6 · Cited by 57 · see citing works
Marginal and Nested Structural Models Using Instrumental Variables
Tan, Z. (2010) · Journal of the American Statistical Association · doi:10.1198/jasa.2009.tm08299 · Cited by 55 · see citing works
Mendelian randomization analysis of case‐control data using structural mean models
Bowden, J., & Vansteelandt, S. (2010) · Statistics in Medicine · doi:10.1002/sim.4138 · Cited by 55 · see citing works
McKeigue, P., Campbell, H. J., Wild, S. H., Vitart, V., Hayward, C., Rudan, I., Wright, A. F., & Wilson, J. F. (2010) · International Journal of Epidemiology · doi:10.1093/ije/dyp397 · Cited by 45 · see citing works
Burgess, S., & Thompson, S. G. (2010) · Statistics in Medicine · doi:10.1002/sim.3843 · Cited by 25 · see citing works
Self-citationPesaran, M. H., Bowden, R. J. J., & Turkington, D. A. (1986) · Economica · doi:10.2307/2553961 · Cited by 24 · see citing works
Hardin, J. W., & Carroll, R. J. (2003) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0400300404 · Cited by 16 · see citing works
Hardin, J. W., & Carroll, R. J. (2003) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0300300404 · Cited by 5 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Mendelian Randomization Analysis With Multiple Genetic Variants Using Summarized Data
Burgess, S., Butterworth, A., & Thompson, S. G. (2013) · Genetic Epidemiology · doi:10.1002/gepi.21758 · Cited by 6,050 · see citing works
Self-citationPierce, B. L., & Burgess, S. (2013) · American Journal of Epidemiology · doi:10.1093/aje/kwt084 · Cited by 1,389 · see citing works
Self-citationBurgess, S., Dudbridge, F., & Thompson, S. G. (2016) · Statistics in Medicine · doi:10.1002/sim.6835 · Cited by 1,153 · see citing works
Self-citationMendelian randomisation for mediation analysis: current methods and challenges for implementation
Carter, A. R., Sanderson, E., Hammerton, G., Richmond, R. C., Davey Smith, G., Heron, J., Taylor, A. E., Davies, N. M., & Howe, L. D. (2021) · European Journal of Epidemiology · doi:10.1007/s10654-021-00757-1 · Cited by 1,215 · see citing works
Burgess, S., Daniel, R. M., Butterworth, A. S., Thompson, S. G., & the EPIC-InterAct Consortium (2015) · International Journal of Epidemiology · doi:10.1093/ije/dyu176 · Cited by 543 · see citing works
Self-citationUse of allele scores as instrumental variables for Mendelian randomization
Burgess, S., & Thompson, S. G. (2013) · International Journal of Epidemiology · doi:10.1093/ije/dyt093 · Cited by 452 · see citing works
Self-citation
- Authors
- Burgess, S., & Thompson, S. G.
- Affiliations
- University of Cambridge (GB)
- Corresponding author
- Burgess, Stephen
- Venue
- Statistics in Medicine
- Publisher
- Wiley
- Published
- 2012-02-29
- Volume
- 31
- Issue
- 15
- Pages
- 1582-1600
- ISSN
- 0277-6715 (ISSN-L)
- 1097-0258
- OpenAlex
- W1997254444
- Cited by
- 87
- References
- 65
- Citations by year
- 2026: 2 · 2025: 9 · 2024: 16 · 2023: 8 · 2022: 5 · 2021: 5 · 2020: 5 · 2019: 6 · 2018: 6 · 2017: 5 · 2016: 4 · 2015: 2 · 2014: 7 · 2013: 5 · 2012: 2
- FWCI
- 4.89
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- Statistics in Medicine (publishedVersion)
- PubMed (publishedVersion)
- Funders
- British Heart Foundation (RG/08/014/24067)
- Medical Research Council
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Keywords
- weak instruments, instrumental variables, two-stage least squares, bias mitigation, coverage properties, binary outcomes, continuous outcomes, Mendelian randomization, uncertainty quantification
- MeSH terms
- Bayes Theorem, Data Interpretation, Statistical, Humans, Risk Factors, Bias, Causality, Confounding Factors, Epidemiologic, Mendelian Randomization Analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
