2018
Direct likelihood-based inference for discretely observed stochastic compartmental models of infectious disease
Ho L, Crawford F, Suchard M. Direct likelihood-based inference for discretely observed stochastic compartmental models of infectious disease. The Annals Of Applied Statistics 2018, 12: 1993-2021. DOI: 10.1214/18-aoas1141.Peer-Reviewed Original ResearchHamiltonian Monte CarloStochastic compartmental modelMonte CarloLikelihood-based inferenceSequential Monte CarloCompartmental modelObserved dataMathematical foundationHigh computational costLoss of accuracyLikelihood evaluationBayesian inferenceSIR modelBroad classRecursion methodComputational costRemoved (SIR) modelEfficient algorithmTransition probabilitiesCarloModel assumptionsInfectious disease epidemicsInferenceCentury plagueApproximationComputational methods for birth‐death processes
Crawford FW, Ho LST, Suchard MA. Computational methods for birth‐death processes. Wiley Interdisciplinary Reviews Computational Statistics 2018, 10 PMID: 29942419, PMCID: PMC6014701, DOI: 10.1002/wics.1423.Peer-Reviewed Original ResearchBirth-death processStatistical inferenceGeneral birth–death processesFinite-time transitionBasic mathematical theoryNon-negative integersContinuous-time Markov chainMaximum likelihood estimationMathematical theoryTheoretical propertiesComputational difficultiesProbability distributionMarkov chainAnalytic expressionsEM algorithmEquilibrium probabilityStatistical workLikelihood estimationSimple caseLinear processRich varietyComputational methodsSimple linear processSummary statisticsInferenceConfidence intervals for linear unbiased estimators under constrained dependence
Aronow P, Crawford F, Zubizarreta J. Confidence intervals for linear unbiased estimators under constrained dependence. Electronic Journal Of Statistics 2018, 12: 2238-2252. DOI: 10.1214/18-ejs1448.Peer-Reviewed Original ResearchLinear unbiased estimatorUnbiased estimatorWald-type confidence intervalsCentral limit theoremInteger linear programConservative coverageLimit theoremAlternative boundsHomoskedasticity assumptionLinear programSuch graphsDependency graphVariance estimatorIndependence relationsEstimatorGraphConstraintsTheoremBoundsDependent outcomesInferenceConfidence intervals
2015
Sex, lies and self-reported counts: Bayesian mixture models for heaping in longitudinal count data via birth–death processes
Crawford FW, Weiss RE, Suchard MA. Sex, lies and self-reported counts: Bayesian mixture models for heaping in longitudinal count data via birth–death processes. The Annals Of Applied Statistics 2015, 9: 572-596. PMID: 26500711, PMCID: PMC4617556, DOI: 10.1214/15-aoas809.Peer-Reviewed Original ResearchLongitudinal count dataImportant statistical problemSelf-reported countsBayesian mixture modelBirth-death processStatistical problemsBayesian hierarchical modelTrue distributionCount dataMixture modelInferential tasksHierarchical modelHeaping processParametersMultiplesModelInferenceEstimationDistributionGridProblemError