2018
There is no impact of exposure measurement error on latency estimation in linear models
Peskoe SB, Spiegelman D, Wang M. There is no impact of exposure measurement error on latency estimation in linear models. Statistics In Medicine 2018, 38: 1245-1261. PMID: 30515870, PMCID: PMC6542365, DOI: 10.1002/sim.8038.Peer-Reviewed Original ResearchConceptsMeasurement error modelLinear measurement error modelsLeast squares estimatorStandard measurement error modelLinear modelError modelRegression coefficient estimatesLikelihood-based methodsMeasurement errorExposure measurement errorSquares estimatorWide classGeneralized linear modelMean functionStatistical modelCovariance structureError settingsNaive estimatorBody mass indexBehavioral risk factorsLatency parametersExposure-disease relationshipsPrimary disease modelTime-varying exposureCoefficient estimates
2004
Equivalence of conditional and marginal regression models for clustered and longitudinal data
Ritz J, Spiegelman D. Equivalence of conditional and marginal regression models for clustered and longitudinal data. Statistical Methods In Medical Research 2004, 13: 309-323. DOI: 10.1191/0962280204sm368ra.Peer-Reviewed Original ResearchGeneral link functionLink functionPositive stable frailtyMarginal regression parametersCertain statistical modelsMarginal mean parametersConditional meanFurther distributional assumptionsNormal random effectsRandom effectsMarginal regression modelsConditional mean functionSlope parameterMarginal meansRegression parametersMean functionStatistical modelProbit linkMeans coincideDistributional assumptionsMean parametersMarginal hazardCovariates of interestEquivalent interpretationWeibull proportional hazards