Modelling a binary response of irvingia cuttings to a rooting hormone
Abstract
Modelling is a statistical method using mathematical
equation to describe the relationship between one or more
dependent variables and a set of explanatory variables
which can be experimental conditions.
In this project, the logistic and Cox regression models are
used to study the effects of the mother trees and hormone
doses on the probabilities and rates of rooting of Irvingia
cuttings. Chapter I gives a general introduction on model
based analysis and some background information on
vegetative propagation. The problematic, the objectives and
the significance of the study are then stated. Reasons for
the used of logistic and Cox model are presented in section
1. 5. In section 1. 6, previous studies on Irvingia
gabonensis add on the application of logistic or Cox
regression models to the analysis of experimental data are
reviewed. Chapter II describes the statistical methodology
used in the study. For the two mocfe'"1s't-h;e' procedures for
parameters estimation, significance tests and checking
model adequacy are presented. The different postulated
models for the Irvingia cuttings data are defined and the
methods for selecting the most appropriate presented. In
chapter III, the main results of the data analysis are
presented, interpreted and commented. The estimated
parameters of the appropriate model corresponding to the
effects of different treatments are given. Then the
predicted probabilities of rooting and the relative rooting
rates for each treatment are presented and compared. The
last chapter deals with the concluding remarks based on the
important results obtained and discuss how these findings
can be beneficial in the development of a practical
methodology of vegetative propagation of Irvingia
gabonensis. Some recommendations are also made for future
study in the area.
Citation
Master of Science degree in BiometrySponsorhip
University of NairobiPublisher
College of Biological & Physical Sciences