A Comparison of Methods of Fitting Several Models to Nutritional Response Data

¹ Department of Agricultural and Applied Economics, The University of Georgia, Athens GA, 30602-2772 U.S.A.
² Department of Poultry Science, The University of Georgia, Athens GA, 30602-2772 U.S.A.

^* To whom correspondence should be addressed. E-mail: gpesti{at}uga.edu.

	Abstract

A variety of models have been proposed to fit nutritional input/output response data. The models are typically nonlinear and therefore fitting the models usually requires sophisticated statistical software and training to use the software. An alternative tool for fitting nutritional response models was developed using widely available and easier-to-use Microsoft Excel software. The tool, implemented as an Excel workbook (NRM.xls), allows simultaneous fitting and side-by-side comparisons of several popular models. This study compares the results produced by the developed tool and Proc NLIN of SAS 9.1. The models compared were the Broken Line (ascending linear and quadratic segments), Saturation Kinetics, four-parameter Logistics, Sigmoidal and Exponential models. The NRM.xls workbook provided nearly identical results to Proc NLIN of SAS 9.1. Furthermore, the workbook successfully fit several models that failed to converge in Proc NLIN of SAS 9.1. Two data sets were used as examples to compare fits by the several models. The results suggest that no particular non-linear model is necessarily best for all nutritional response data.

Key Words: nutritional requirements, mathematical modeling