A STUDY ON ROW-COLUMN DESIGNS
KrishiKosh
View Archive InfoField | Value | |
Title |
A STUDY ON ROW-COLUMN DESIGNS
|
|
Creator |
KALLOL SARKAR
|
|
Contributor |
. Cini Varghese
|
|
Subject |
economics, technological changes, costs, manpower, marketing, forestry, byproducts, irrigation, horticulture, crops
|
|
Description |
RFT-3173
When the heterogeneity present in the experimental material is from two cross classified sources, then two-dimensional blocking of the experimental units is recommended for control or reduction of experimental error. The two blocking systems are referred to as row blocking and column blocking and the resulting designs are known as row-column designs. These designs are widely used in agronomic or forestry trials where a large number of hybrid varieties are to be compared leading to a large experimental area and hence, fertility trends may exist in the land in two perpendicular directions. If an experimenter wants to study the effect of two or more factors simultaneously in a row-column setup, adopting an appropriate factorial row-column design is statistically more informative and efficient than running separate experiments for each factor. Some families of symmetric/asymmetric factorial row-column designs in complete/incomplete rows/columns have been obtained. Further, there are many experimental situations in which the experimenter wants to compare a set of new (test) treatments with an already existing (control) treatment giving more emphasis to test vs. control comparisons. Some general methods of constructing balanced treatment-control row-column designs in complete/incomplete rows/columns have been developed. Further, a class of structurally incomplete balanced treatment-control rowcolumn designs has also been obtained where treatments are applied to a subset of the available experimental units. Need for such designs is apparent when the blocking criteria are implemented in sequence where the number of treatments is lesser in the second stage than the first stage. Programs have been written in PROC IML of SAS software for computing the information matrix and variances pertaining to different groups of treatment comparisons in factorial as well as balanced treatment-control row-column design setup. List of parameters of designs constructed for a practical range of parameters has been prepared along with variance of estimated contrasts pertaining to different treatment comparisons. |
|
Date |
2016-12-15T14:27:03Z
2016-12-15T14:27:03Z 2011 |
|
Type |
Thesis
|
|
Identifier |
http://krishikosh.egranth.ac.in/handle/1/90329
|
|
Format |
application/pdf
|
|
Publisher |
INDIAN AGRICULTURAL STATISTICS RESEARCH INSTITUTE
|
|