Fast stable relative risk regression using an overparameterised EM algorithm

Mark W. Donoghoe, Ian C. Marschner

    Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

    Abstract

    Relative risk regression models can be fitted using a log-link binomial GLM, however standard algorithms can suffer convergence problems. Combinatorial EM (CEM) algorithms that provide stable convergence can be computationally intensive, particularly for large models. We present a new approach using an EM algorithm with an overparameterised model that retains the stability of the CEM algorithm but greatly reduces computing time. This is demonstrated with a small example in which modified Fisher scoring fails to converge to the MLE, and a bootstrap analysis of data from a clinical trial in heart attack patients.
    Original languageEnglish
    Title of host publicationProceedings of the 31st International Workshop on Statistical Modelling
    EditorsJean-François Dupuy, Julie Josse
    Place of PublicationAmsterdam, the Netherlands
    PublisherStatistical Modelling Society
    Pages93-98
    Number of pages6
    Volume1
    Publication statusPublished - 2016
    EventInternational Workshop on Statistical Modelling (31st : 2016) - Rennes, France
    Duration: 4 Jul 20168 Jul 2016
    Conference number: 31st

    Conference

    ConferenceInternational Workshop on Statistical Modelling (31st : 2016)
    Abbreviated titleIWSM2016
    CountryFrance
    CityRennes
    Period4/07/168/07/16

    Keywords

    • Binomial regression
    • EM algorithm
    • Relative risk

    Cite this