Premium
Prediction of Growth: A Hierarchical Bayesian Approach
Author(s) -
Arjas Elja,
Liu Liping,
Maglaperidze Niko
Publication year - 1997
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.4710390612
Subject(s) - markov chain monte carlo , hierarchy , bayesian probability , nonparametric statistics , growth curve (statistics) , computer science , markov chain , hierarchical database model , variable order bayesian network , approximate bayesian computation , computation , flexibility (engineering) , econometrics , bayesian inference , mathematics , artificial intelligence , algorithm , machine learning , data mining , statistics , inference , economics , market economy
A nonparametric hierarchical growth curve model is proposed. Different levels in the model hierarchy are intended to correspond to different sources of variation in an individual's growth. The nonparametric character of the model offers considerable flexibility in fitting the growth curves to empirical data. Here the emphasis is on prediction, and for this purpose the adopted Bayesian inferential approach seems particularly natural and efficient. A Markov chain Carlo method is used to perform the numerical computations. As an illustration of the techniques, we consider the growth of children, during their first two years.