Comparative analysis on selecting growth function based on three different information criteria for the purpose of carbon estimation

Abstract

Growth models are in general constructed by some kind of mathematical function, so-called "growth function". Since there are several possible growth functions, the first task for the growth analysis is to select the most suitable growth function for the target data among candidates. In this paper, three statistical procedures are presented and compared on the basis of the information criterion for the purpose of selecting a growth function. A demonstrative example is provided to show how the three statistical procedures differ and select different growth functions for the same growth data, and how differently the amount of carbon sequestrated in a forest stand is affected by the selected growth function. © 2013 Copyright Taylor and Francis Group, LLC.