• RAPID PARAMETER ESTIMATION OF THREE PARAMETER NONLINEAR GROWTH MODELS

Abstract

The aim of this study is to determine a suitable method to estimate the parameters of some non-linear growth models using various methods of estimation. Six different estimation methods are used for estimating the parameters of the Monomolecular, Gompertz and Logistic growth models. Three different forestry data sets are used for testing the validity of the methods. The best fit method is selected on the basis of root mean square error (RMSE). It is found that the first four methods of estimation used in this paper are very useful to specify the initial values in the estimation of the parameters using any iterative procedures. It is noted that the most suitable growth model for the top height growth data is the Gompertz model which may be obtained by estimating the parameters using the Newton-Raphson method under the assumption that the rate parameter is known. For the same method, the Gompertz growth model also gives the proper explanation for cumulative basal area production. The Logistic growth model is found to be most suitable for the mean diameter at breast height with the Composite method assuming that the rate parameter is known from method of three partial sums.