Nonlinear growth models for modeling oil palm yield growth
This study provided the basic needs of parameters estimation for nonlinear growth model such as partial derivatives of each model, determination of initial values for each parameter and statistical tests of industrial usage. Twelve nonlinear growth models and its partial derivatives for oil palm y...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Science Publications
2005
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/2911/ http://eprints.utm.my/2911/1/AzmeKhamis2005_NonlinerGrowthModelsForModeling.pdf |
| Summary: | This study provided the basic needs of parameters estimation for nonlinear growth model such as
partial derivatives of each model, determination of initial values for each parameter and statistical tests of
industrial usage. Twelve nonlinear growth models and its partial derivatives for oil palm yield growth are
presented in this study. The parameters are estimated using the Marquardt iterative method of nonlinear
regression relating oil palm yield growth data. The best model was selected based on the model
performance and it can be used to estimate the oil palm yield at any age of oil palm. This study found that
the Gompertz, logistic, log-logistic, Morgan-Mercer-Flodin and Chapman-Richard growth models have the
ability for quantifying a growth phenomenon that exhibit a sigmoid pattern over time. Based on the
statistical testing and goodness of fit, the best model is the Logistic model and followed by the Gompertz
model, Morgan-Mercer-Flodin, Chapman-Richard (with initial stage) and Log-logistic growth models |
|---|