Show simple item record

dc.contributor.authorDavies, Neil Men
dc.contributor.authorvon, Hinke Kessler Scholder Stephanieen
dc.contributor.authorFarbmacher, Helmuten
dc.contributor.authorBurgess, Stephenen
dc.contributor.authorWindmeijer, Franken
dc.contributor.authorDavey, Smith Georgeen
dc.date.accessioned2014-10-22T16:23:05Z
dc.date.available2014-10-22T16:23:05Z
dc.date.issued2014-11-10en
dc.identifier.citationStatistics in Medicine 34 (3): 454-68. doi: 10.1002/sim.6358en
dc.identifier.issn0277-6715
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/246246
dc.description.abstractInstrumental variable estimates of causal effects can be biased when using many instruments that are only weakly associated with the exposure. We describe several techniques to reduce this bias and estimate corrected standard errors. We present our findings using a simulation study and an empirical application. For the latter, we estimate the effect of height on lung function, using genetic variants as instruments for height. Our simulation study demonstrates that, using many weak individual variants, two-stage least squares (2SLS) is biased, whereas the limited information maximum likelihood (LIML) and the continuously updating estimator (CUE) are unbiased and have accurate rejection frequencies when standard errors are corrected for the presences of many weak instruments. Our illustrative empirical example uses data on 3,631 children from England. We used 180 genetic variants as instruments, and compare conventional ordinary least squares (OLS) estimates to results for the 2SLS, LIML and CUE instrumental variable estimates using an unweighted or weighted allele score as single instruments. In conclusion, the allele scores and CUE gave consistent estimates of the causal effect. In our empirical example, estimates using the allele score were more efficient. CUE with corrected standard errors, however, provides a useful additional statistical tool in applications with many weak instruments. The CUE may be preferred over an allele score if the population weights for the allele score are unknown, or when the causal effects of multiple risk factors are estimated jointly.
dc.languageen_USen
dc.language.isoen_USen
dc.publisherWiley
dc.titleThe many weak instruments problem and Mendelian randomizationen
dc.typeArticle
dc.description.versionThis is the final version. It was first published by Wiley at onlinelibrary.wiley.com/doi/10.1002/sim.6358/abstracten
prism.endingPage468
prism.publicationDate2014en
prism.publicationNameStatistics in Medicineen
prism.startingPage454
prism.volume34en
dc.rioxxterms.funderMedical Research Council
dc.rioxxterms.funderMRC
dc.rioxxterms.funderMRC
dc.rioxxterms.funderWellcome Trust
dc.rioxxterms.funderERC
dc.rioxxterms.projectidG0600705
dc.rioxxterms.projectidG1002345
dc.rioxxterms.projectidMC_UU_12013/1-9
dc.rioxxterms.projectid100114
dc.rioxxterms.projectidDevHEALTH 269874
dcterms.dateAccepted2014-10-20en
rioxxterms.versionofrecord10.1002/sim.6358en
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserveden
rioxxterms.licenseref.startdate2014-11-10en
dc.contributor.orcidBurgess, Stephen [0000-0001-5365-8760]
dc.identifier.eissn1097-0258
rioxxterms.typeJournal Article/Reviewen
pubs.funder-project-idMRC (MR/L003120/1)
pubs.funder-project-idWellcome Trust (100114/Z/12/Z)
pubs.funder-project-idBritish Heart Foundation (RG/08/014/24067)


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record