Seminario:"A Bayesian multivariate growth curve model"

Seminario:"A Bayesian multivariate growth curve model"

Fecha: 22-11-2013

Autor: Emmanuel Lesaffre (Department of Biostatistics, Erasmus MC, Rotterdam, the Netherlands y I-Biostat, KU Leuven, Leuven, Belgium)

Título:”A Bayesian multivariate growth curve model”

Lugar: Facultad de Ciencias Económicas y Empresariales (Bilbao)



A Bayesian multivariate growth curve model

Emmanuel Lesaffre
Department of Biostatistics, Erasmus MC, Rotterdam, the Netherlands
I-Biostat, KU Leuven, Leuven, Belgium

Most longitudinal growth curve models evaluate the evolution of each of the body characteristics
separately. When applied to a `normal population’ this exercise leads to univariate reference curves
against which newly included individuals can be evaluated. However, normal growth should be
evaluated in totality, i.e. by evaluating all body characteristics jointly. Recently, Cole et al. suggested
the SITAR model which expresses individual growth curves by three subject-specific parameters
indicating their deviation from a flexible overall growth curve. This model allows characterizing
normal growth in a flexible though compact manner. In this paper we generalize the SITAR model in a
Bayesian way to multiple dimensions. The multivariate SITAR model allows for multivariate reference
regions and allows for better conditional prediction, given the past growth. The usefulness of the
model is illustrated on measurements of embryonic growth taken in the first semester of pregnancy,
collected in an ongoing study at Erasmus MC, Rotterdam. Further, we demonstrate how the model
can be used to find determinants of early growth.

en Uncategorized