Naslov
MAIN EFFECT META PRINCIPAL COMPONENT ANALYSIS (ME-METAPCA) OF PLANT GROWTH REGULATOR TREATMENT EFFECT ON SIMULATED MULTIPLE APPLE DATA

Autori
Borut Bosančić, Marija Pecina, Nikola Mićić

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
AgroReS 2016 Book of Abstracts / Đurić Gordana - Banja Luka : Poljoprivredni fakultet Univerziteta u Banjoj Luci, 2016, 51-51

ISBN
978-99938-93-37-0

Skup
5th International Symposium on Agricultural Sciences AgroReS 2016

Mjesto i datum
Banja Luka, Bosna i Hercegovina, 29.02.2016. - 03.03.2016

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Meta-Analysis, PCA, Multivariate, Biometrics, Fruit Characteristics

Sažetak
Meta-Analysis as a statistical and analytical method for combining and synthesizing several independent studies and integrating results into a common result and conclusion has not been established in agricultural research yet. Moreover, combination of Meta-Analysis and Principal Components Analysis of main treatment effects (ME-MetaPCA) is a novel approach for analysis of experimental treatment effects. The aim of this paper is to introduce this approach through agricultural model, where it is required, in order to objectively and effectively summarize and generalize conclusions through multiple researches with multiple variables. Simulated data are modeled as real multiple fruit characteristics that define both yield quantity and fruit quality in apple, which is the case in most studies of agricultural crops. Treatment generally affects several fruit characteristics and its effect typically varies throughout different studies and varieties. Finding the real underlying treatment effect size and grouping the studied varieties accordingly would be of practical use for both agricultural researcher and producer. The simulated data were modeled as Plant Growth Regulator (PGR) treatment in several studies where multiple apple varieties were treated and multiple fruit characteristics measured. Results are displayed in form of forest plots related to Meta-Analysis for individual characteristics followed by graphical presentation in principal component space of main effect’s eigenvectors of measured characteristics in studied varieties. This leads to better and more objective understanding of the general rules regarding the effect of the PGR treatment and its influence over various measured fruit characteristics and studied varieties, their grouping and dispersion. Key Words: Meta-Analysis, PCA, Multivariate, Biometrics, Fruit Characteristics

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Poljoprivreda (agronomija)

POVEZANOST RADA