The genetic study of certain quantitative traits in growth curves as a function of time has recently been of major scientific interest to explore the developmental evolution processes of biological subjects. Various parametric approaches in the statistical literature have been proposed to study the quantitative-trait-loci (QTL) mapping of the growth curves as multivariate outcomes. In this article, we view the growth curves as functional quantitative traits and propose some semiparametric models to relax the strong parametric assumptions which may not be always practical in reality. Appropriate inference procedures are developed to estimate the parameters of interest which characterise the possible QTLs of the growth curves in the models. Recently developed multiple comparison testing procedures are applied to locate the statistically meaningful QTLs. Numerical examples are presented with simulation studies and analysis of real data.


Genetics | Statistical Methodology | Statistical Models | Statistical Theory | Survival Analysis