Articles
MODEL SELECTION FOR GROWTH PATTERN OF THE APPLE FRUIT: BILINEAR MODEL FAVORED OVER OTHER NONLINEAR GROWTH MODELS
Article number
803_37
Pages
293 – 298
Language
English
Abstract
Development of mathematical models for growth simulation and yield prediction would be helpful for improvement of crop management.
Fruit growth especially represents an important process because it is directly related to crop yield and so with the economical productivity of the crop.
In this survey seven nonlinear growth models including a recently introduced bilinear (Linearized biexponential, LinBiExp) model were used for modeling the inherent growth pattern of apple fruit on the basis of fruit length, diameter and volume.
The parameters were estimated using the Marquardt iterative method of nonlinear regression relating to apple fruit growth data.
Selection of the most adequate model to describe a given data set requires the use of quantitative model selection criteria, such as partial F-test, the Akaike information criterion and the Schwarz Bayesian information criterion which are suitable for comparing differently parameterized models in terms of the quality and strength of the fit.
The study found that the bilinear model was more suitable than the other nonlinear growth models for fitting the apple growth data according to the measured indices (fruit diameter and volume).
Fruit growth especially represents an important process because it is directly related to crop yield and so with the economical productivity of the crop.
In this survey seven nonlinear growth models including a recently introduced bilinear (Linearized biexponential, LinBiExp) model were used for modeling the inherent growth pattern of apple fruit on the basis of fruit length, diameter and volume.
The parameters were estimated using the Marquardt iterative method of nonlinear regression relating to apple fruit growth data.
Selection of the most adequate model to describe a given data set requires the use of quantitative model selection criteria, such as partial F-test, the Akaike information criterion and the Schwarz Bayesian information criterion which are suitable for comparing differently parameterized models in terms of the quality and strength of the fit.
The study found that the bilinear model was more suitable than the other nonlinear growth models for fitting the apple growth data according to the measured indices (fruit diameter and volume).
Authors
A. Saei, Z. Zamani, A. Talaie, M.R. Fatahi
Keywords
Malus domestica L., nonlinear growth model, fruit growth, model selection criteria
Online Articles (41)