Fitting and Comparing Several Growth Curves of the Generalized von Bertalanffy Type

Abstract

The earlier version of the von Bertalanffy growth formula (VBGF) has three parameters and is a special case of the "generalized" VBGF which has four parameters. The special VBGF can easily lead to erroneous results whereas the large flexibility built into the generalized VBGF enables it to provide a biologically realistic fit to the data. I present a statistical method and a computer program in Fortran (1) to fit the generalized VBGF to a sample (a sample comprises several age groups, with one or more observations in an age group), (2) to examine statistically the fit to a sample, and (3) to compare statistically these curves (and their individual parameters) fitted to several samples. An example is provided, which also compares the fit of the generalized VBGF with the special VBGF.