ESTIMATION OF GROWTH PARAMETERS USING NON-LINEAR MIXED MODEL AND COMPARISON WITH FIXED EFFECT MODEL IN ANIMALS
Loading...
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
INDIAN AGRICULTURAL STATISTICS RESEARCH INSTITUTE INDIAN AGRICULTURAL RESEARCH INSTITUTE NEW DELHI
Abstract
Modeling growth of an animal is a complex process, because it requires describing
longitudinal measurements with few parameters with biological interpretation. With
longitudinal data, the variance of observations may increase with time (age), and repeated
measurements of an individual over time are correlated. The non-independence of data
violates a key assumption underlying many statistical procedures and has been ignored in
most traditional non-linear fixed effect models. A solution to this problem is the use of
non-linear mixed effect models (NLMM). Mixed effects models for longitudinal data such
as growth have become of interest because of their flexible covariance structure and their
ability to handle unbalanced data. A NLMM makes it possible to account for random
covariates before testing for fixed effects and control autocorrelation in repeated
measures. Nonlinear mixed effect models also account for multiple sources of
heterogeneity in data through the inclusion of random effects in the models.
In this study, growth data of Goat and Pig has been used. Attempt has been made to
develop the Von-bertalanffy mixed model. Logistic, Gompertz and Von-bertalanffy fixed
and mixed models have been explored for these data. Comparison of the models i.e.
between fixed and mixed type of the same model and among different fixed and mixed
models has been attempted. The goodness of fit statistics like i.e. Mean Square Error
(MSE) and Root Mean Square Error (RMSE) of the fitted models has been computed. The
parameters of the best fitted models along with their corresponding standard error are
estimated. It has been found that Logistic mixed effect model is performed best for the
Goat and Pig data. The prediction of this model is also better. Computer programs in SAS
and R have been developed for the data analysis. In literature it has been found that
Logistic mixed model is most appropriate for animals. So it may be concluded that the
logistic mixed effect model can be used for other animal.
Description
T-9251
Keywords
null