Nonparametric Modelling of Longitudinal Data: A Varying Coefficients Model

  1. Susan Orbe-Mandaluniz 2
  2. Vicente Núñez-Antón 2
  3. Juan M. Rodríguez-Póo 1
  1. 1 Universidad de Zaragoza
    info

    Universidad de Zaragoza

    Zaragoza, España

    ROR https://ror.org/012a91z28

  2. 2 Universidad del País Vasco/Euskal Herriko Unibertsitatea
    info

    Universidad del País Vasco/Euskal Herriko Unibertsitatea

    Lejona, España

    ROR https://ror.org/000xsnr85

Actas:
Statistical Modelling: Proceedings of the 19th International Workshop on Statistical Modelling. Florence, Italy, July 4-8, 2004

Editorial: Firenze University Press

ISBN: 88-8453-193-4

Año de publicación: 2004

Páginas: 234-238

Tipo: Aportación congreso

Resumen

We propose a very flexible and general semiparametric model that allows for time-varying coefficients and/or covariate-varying (including groupsvarying or subjects-varying) coefficients in a longitudinal data setting. Tests for model specification are proposed and, thus, this proposal allows to discriminate between the different sources of variation for the regression coefficients. The model is applied to several longitudinal data examples and, in addition, its performance is studied when compared to other more restricted proposals.