Muhammad Aman Ullah

Study of correlation among yield related traits and path coefficient analysis in rice (Oryza sativa L.)

In order to determine the associations among yield components and their direct and indirect influence on grain yield of rice, an experiment was conducted. For this purpose, 30 genotypes collected from different sources were tested in a randomized complete block design. The phenotypic correlations among the yield traits and their path coefficient were estimated. Grain yield was significantly correlated with its component characters; number of productive tillers per plant, number of grains per panicle and flag leaf area. Path analysis revealed that days to maturity had the highest direct effect (0.751) on grain yield per plant. In addition, the yield components had positive direct effect on grain yield except the days to heading (-0.834). The order of yield components was the number of productive tillers per plant, flag leaf area and 1000 grain weight. The improvement in grain yield will be efficient if the selection is based on biological yield components, number of productive tillers per plant and flag leaf area. These traits may also be utilized in pure line selection. Keywords : Rice ( Oryza sativa ), path coefficient, phenotypic correlation, yield components

Assessing Influence on the Liu Estimates in Linear Regression Models

The Liu estimator has been developed as an alternative to the ordinary least squares estimator in the presence of collinearity among the elements of regressors in linear regression models. We present the DFFITS and different versions of the Cook distance analogous to the ones given for the ordinary linear regression models of each individual observation on the Liu estimates. We suggest a version of the Cook distance based on one-step approximation. The mean shift outlier model for the Liu regression has also been investigated. Moreover, using the Sherman-Morrison-Woodbury theorem, we find approximate versions of the DFFITS and the Cook distance. The proposed diagnostics are evaluated on two data sets and yield promising results.