Fitting joint models of longitudinal observations and time to event by sequential Bayesian updating

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

Joint modelling of longitudinal measurements and time to event, with longitudinal and event submodels coupled by latent state variables, has wide application in biostatistics. Standard methods for fitting these models require numerical integration to marginalize over the trajectories of the latent states, which is computationally prohibitive for high-dimensional data and for the large data sets that are generated from electronic health records. This paper describes an alternative model-fitting approach based on sequential Bayesian updating, which allows the likelihood to be factorized as the product of the likelihoods of a state-space model and a Poisson regression model. Updates for linear Gaussian state-space models can be efficiently generated with a Kalman filter and the approach can be implemented with existing software. An application to a publicly available data set is demonstrated.

Original languageEnglish
Pages (from-to)1934-1941
Number of pages8
JournalStatistical Methods in Medical Research
Volume31
Issue number10
Early online date31 May 2022
DOIs
Publication statusPublished - 1 Oct 2022

Keywords / Materials (for Non-textual outputs)

  • Bayes Theorem
  • Biometry/methods
  • Biostatistics
  • Linear Models
  • Longitudinal Studies
  • Models, Statistical
  • Normal Distribution

Fingerprint

Dive into the research topics of 'Fitting joint models of longitudinal observations and time to event by sequential Bayesian updating'. Together they form a unique fingerprint.

Cite this