Statistical Models for Prediction of Height and Diameter Relationships in Chir Pine Trees in Jammu
Loading...
Date
2020-12
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Sher-e-Kashmir University of Agricultural Sciences and Technology Jammu, J&K
Abstract
An investigation entitled on “Statistical Models for Prediction of Height and Diameter Relationships in Chir Pine Trees in Jammu” was conducted on secondary data of 300 Chir Pine trees of Jammu on height and diameter variables (100 trees per forest division i.e Jammu, Nowshera, Batote) from JK forest department, and accordingly data was fitted on eleven height diameter models. The aim of this study was to fit various height diameter statistical models for studying tree height and diameter relationships and to evaluate these fitted models in terms of their predictive performances. The eleven height diameter models were fitted on data of three forest divisions. Almost all the height diameter models fitted resulted in significant coefficients, which indicated that these models were capturing the height diameter relationships of Chir trees well. Apart from this various selection criteria like RSE, RMSE, MAE, AIC, BIC, R2, Adj. R2 and Shapiro wilk test were used to study the performance of the fitted models and normality of errors. The results of these criteria were generated from various libraries of R studio (version 3.5.1, 2018), besides various function of R software were also created in this study. The fitted models were further validated by cross validation technique in which whole data set was splited in a ratio of 80:20 where 80 percent observations were randomly allocated to training data set and remaining 20 percent in testing data. Based on the results of validation MG (Manfred N3) and MJ (Michaelis-Menten2) came out to be best fitted models in comparison to other models used in this study, as both these models accounted for the utmost fraction of total height variations, and more crucially with lowest prediction error rate and both models appeared to be biologically more realistic.
Description
Keywords
Citation
Preferred for your work_ Statistical models for prediction of height and diameter relationships in chir pine trees in jammu, Division of Stat & com (FBSc) (2020).