Statistical modeling for analyzing grain yield of durum wheat under rainfed conditions in Azad Jammu Kashmir, Pakistan

Abstract One of the most important traits that plant breeders aim to improve is grain yield which a highly quantitative trait controlled by various agro-morphological traits. Twelve morphological such as Germination Percentage, Days Spike Emergence, Plant Height, Length, Awn Tillers/Plant, Leaf Angle, Seeds/Spike, Thickness, 1000-Grain Weight, Harvest Index and Maturity have been considered independent factors. Correlation, regression, principal component analysis (PCA) are used identify different durum wheat traits, significantly contribute yield. The necessary assumptions required for applying regression modeling tested all satisfied observed data. outliers detected in observations fixed Grain Yield. Some but outlying did not show any influence on fit. For selecting parsimonious model wheat, best subset stepwise techniques applied. revealed Seeds/Spike marked increasing effect whereas thickness has negative While identified simple correlation coefficient specified significant positive Yield with Number Index. These results directed importance characters their impact PCA showed variation (70%) among data set can be explained first five components. It also Seeds/Spike; Weight higher contributing Based it recommended these parameters might focused future breeding programs develop high varieties.