Modelling Experimental Mammary Adenocarcinoma Growth, Using Nonlinear Mixed-Effects Models and S-Plus

by Rachid El Halimi, Jordi Ocaņa, M. Carme Ruiz de Villa, Montserrat Solanas,Eduardo Escrich.

Abstract: The process of fitting a non-linear mixed model to longitudinal data can be complex if the data show high inter-subject variability. In order to illustrate all the steps if this process, we analyse the data from an experiment on breast cancer where the growth of the tumour volume is modelled in order to determine the influence of an external factor, diet. S-Plus software has been used, and the most relevant sentences to follow the analysis are described. Some strengths and weaknesses of this software tool are also included in the discussion.

Key Words: Nlme library, repeated measures, residuals heteroscedasticity and correlation

Authors:
M. Carme Ruiz de Villa, emruiz_de_villa@ub.edu
ordi Ocaņa, jocana@ub.edu
Rachid El Halimi,
Montserrat Solanas,
Eduardo Escrich

Editor: Avner Bar-Her,avner@bar-hen.net

READING THE ARTICLE: You can read the article in portable document (.pdf) format (427066 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2305 times since July 24, 2006.


Return to the InterStat Home Page.