No Detail Found!

Analysis of a group of randomized basic designs and latin square design with heteroscedastic model.

Author(s)
Ali Akber Memon Institute of Mathematics & Computer Science, University of Sindh, Jamshoro, Pakistan
Lal Chand Mathematics Department, Shah Abdul Latif University, Khairpur, Pakistan
Azizullah Memon Statistics Department, University of Sindh, Jamshoro, Pakistan
Muhammad Hassan Dhaunroo Basic Engineering Department, Sindh Agricultural University, Tandojam, Pakistan
Abstract
For a group of experiments it is assumed that the error variance is constant for a particular experiment but it varies from experiment to experiment. Assuming error variance to be known, estimators of the treatment parameters are obtained by weighted (generalized) Least Squares Method and the corresponding analysis is given for a group of completely randomized designs, randomized block designs and latin square designs. For unknown error variances, an adjustment of the statistics using estimated weights is proposed for removing much of the resulting bias. In this paper, assuming error variances to be known, the estimates of treatment parameters of a group of basic designs are obtained and the corresponding analysis is provided with the help of the weighted . (generalized) Least Squares Method.
Publication Details
Page(s) 165-175
DOI DOI not available
Published Journal: Scientific Sindh, Volume: 2, Issue: , Year: 1994

Request the full text from PASTIC

0

Citations

0

Downloads

39

Views