Abstract:
Principal Component Analysis (PCA) is one of the oldest and most widely used
multivariate statistical techniques. The original purpose of PCA was to reduce a large
number (p) of variables to a much small number (m) of principal components whilst
retaining as much as possible of the variation in the p-original variables. The technique
is especially useful if m«<p, and if the m principal components can be readily
interpreted.
The main objective of this research study is to find the growth performance of seedlings
of seven rain forest canopy, dominant Shorea species, in Kalutara district within the
humid zone of Sri Lanka for a period of 24 months.
A research was carried out by Prof LA. U.N. Gunetilleka and Prof C. VS. Gunetilleka
(1989) to collect the data from 336 Shorea seedlings which were transplanted at
Kalutara district in 125m altitude elevations. At the end of the experiment period only
112 seedlings were chosen at random for the analysis. The measurements of constituent
parts such as stem weight, leaf weight, tap root weight, fine root weight, leaf number
and total dry mass of those randomly selected seedlings were recorded in the process.
The above two researchers analyzed the recorded data using the ANOVA and GLM
procedure of SAS in one-way ANOVA models for each of the attributes of these
constituent parts, to be measured separately to compare the performance of species
grown at this district.
In this study, all the constituent parts of one species is combined using the principal
component analysis, and used this new combined new variable to compare the
performance of seven species grown at this district. To analyze this combined and
reduced response variable (data) ANOVA procedure of SAS was used. Tu key's
studentized range method was used for multiple comparison of means among species
at the 5% level of significance.
