Model selection curves for survival analysis with accelerated failure time models

J. H. Karami, K. Luo, T. Fung

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

    Abstract

    Many model selection processes involve minimizing a loss function of the data over a set of models. A recent introduced approach is model selection curves, in which the selection criterion is expressed as a function of penalty multiplier in a linear (mixed) model or generalized linear model. In this article, we have adopted the model selection curves for accelerated failure time (AFT) models of survival data. In our simulation study, it was found that for data with small sample size and high proportion of censoring, the model selection curves approach outperformed the traditional model selection criteria, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). In other situations with respect to sample size and proportion of censoring, the model selection curves correctly identify the true model whether it is a full or reduced model. Moreover, through bootstrapping, it was shown that the model selection curves can be used to enhance the model selection process in AFT models.
    LanguageEnglish
    Title of host publication60th ISI World Statistics Congress
    Subtitle of host publicationproceedings
    Place of PublicationThe Hague, The Netherlands
    PublisherInternational Statistical Institute
    Pages2642-2647
    Number of pages6
    ISBN (Print)9789073592353
    Publication statusPublished - 2015
    EventWorld Statistics Congress of the International Statistical Institute (60th : 2015) - Rio de Janeiro
    Duration: 26 Jul 201531 Jul 2015

    Conference

    ConferenceWorld Statistics Congress of the International Statistical Institute (60th : 2015)
    CityRio de Janeiro
    Period26/07/1531/07/15

    Fingerprint

    Accelerated Failure Time Model
    Survival Analysis
    Model Selection
    Curve
    Censoring
    Proportion
    Model Selection Criteria
    Linear Mixed Model
    Bayesian Information Criterion
    Akaike Information Criterion
    Survival Data
    Reduced Model
    Bootstrapping
    Small Sample Size
    Generalized Linear Model
    Loss Function
    Multiplier
    Penalty
    Sample Size
    Simulation Study

    Cite this

    Karami, J. H., Luo, K., & Fung, T. (2015). Model selection curves for survival analysis with accelerated failure time models. In 60th ISI World Statistics Congress: proceedings (pp. 2642-2647). The Hague, The Netherlands: International Statistical Institute.
    Karami, J. H. ; Luo, K. ; Fung, T. / Model selection curves for survival analysis with accelerated failure time models. 60th ISI World Statistics Congress: proceedings. The Hague, The Netherlands : International Statistical Institute, 2015. pp. 2642-2647
    @inproceedings{dd42dce2901541b58d2d490ce17c740b,
    title = "Model selection curves for survival analysis with accelerated failure time models",
    abstract = "Many model selection processes involve minimizing a loss function of the data over a set of models. A recent introduced approach is model selection curves, in which the selection criterion is expressed as a function of penalty multiplier in a linear (mixed) model or generalized linear model. In this article, we have adopted the model selection curves for accelerated failure time (AFT) models of survival data. In our simulation study, it was found that for data with small sample size and high proportion of censoring, the model selection curves approach outperformed the traditional model selection criteria, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). In other situations with respect to sample size and proportion of censoring, the model selection curves correctly identify the true model whether it is a full or reduced model. Moreover, through bootstrapping, it was shown that the model selection curves can be used to enhance the model selection process in AFT models.",
    keywords = "log-likelihood function, penalty function, penalty multiplier, survival data",
    author = "Karami, {J. H.} and K. Luo and T. Fung",
    year = "2015",
    language = "English",
    isbn = "9789073592353",
    pages = "2642--2647",
    booktitle = "60th ISI World Statistics Congress",
    publisher = "International Statistical Institute",
    address = "Netherlands",

    }

    Karami, JH, Luo, K & Fung, T 2015, Model selection curves for survival analysis with accelerated failure time models. in 60th ISI World Statistics Congress: proceedings. International Statistical Institute, The Hague, The Netherlands, pp. 2642-2647, World Statistics Congress of the International Statistical Institute (60th : 2015), Rio de Janeiro, 26/07/15.

    Model selection curves for survival analysis with accelerated failure time models. / Karami, J. H.; Luo, K.; Fung, T.

    60th ISI World Statistics Congress: proceedings. The Hague, The Netherlands : International Statistical Institute, 2015. p. 2642-2647.

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

    TY - GEN

    T1 - Model selection curves for survival analysis with accelerated failure time models

    AU - Karami,J. H.

    AU - Luo,K.

    AU - Fung,T.

    PY - 2015

    Y1 - 2015

    N2 - Many model selection processes involve minimizing a loss function of the data over a set of models. A recent introduced approach is model selection curves, in which the selection criterion is expressed as a function of penalty multiplier in a linear (mixed) model or generalized linear model. In this article, we have adopted the model selection curves for accelerated failure time (AFT) models of survival data. In our simulation study, it was found that for data with small sample size and high proportion of censoring, the model selection curves approach outperformed the traditional model selection criteria, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). In other situations with respect to sample size and proportion of censoring, the model selection curves correctly identify the true model whether it is a full or reduced model. Moreover, through bootstrapping, it was shown that the model selection curves can be used to enhance the model selection process in AFT models.

    AB - Many model selection processes involve minimizing a loss function of the data over a set of models. A recent introduced approach is model selection curves, in which the selection criterion is expressed as a function of penalty multiplier in a linear (mixed) model or generalized linear model. In this article, we have adopted the model selection curves for accelerated failure time (AFT) models of survival data. In our simulation study, it was found that for data with small sample size and high proportion of censoring, the model selection curves approach outperformed the traditional model selection criteria, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). In other situations with respect to sample size and proportion of censoring, the model selection curves correctly identify the true model whether it is a full or reduced model. Moreover, through bootstrapping, it was shown that the model selection curves can be used to enhance the model selection process in AFT models.

    KW - log-likelihood function

    KW - penalty function

    KW - penalty multiplier

    KW - survival data

    M3 - Conference proceeding contribution

    SN - 9789073592353

    SP - 2642

    EP - 2647

    BT - 60th ISI World Statistics Congress

    PB - International Statistical Institute

    CY - The Hague, The Netherlands

    ER -

    Karami JH, Luo K, Fung T. Model selection curves for survival analysis with accelerated failure time models. In 60th ISI World Statistics Congress: proceedings. The Hague, The Netherlands: International Statistical Institute. 2015. p. 2642-2647