Thumbnail Image

Bihar Agricultural University, Sabour

Bihar Agricultural University, Sabour established on 5th August, 2010 is a basic and strategic institution supporting more than 500 researchers and educationist towards imparting education at graduate and post graduate level, conducting basic, strategic, applied and adaptive research activities, ensuring effective transfer of technologies and capacity building of farmers and extension personnel. The university has 6 colleges (5 Agriculture and 1 Horticulture) and 12 research stations spread in 3 agro-ecological zones of Bihar. The University also has 21 KVKS established in 20 of the 25 districts falling under the jurisdiction of the University. The degree programmes of the university and its colleges have been accredited by ICAR in 2015-16. The university is also an ISO 9000:2008 certified organisation with International standard operating protocols for maintaining highest standards in teaching, research, extension and training.VisionThe Bihar Agricultural University was established with the objective of improving quality of life of people of state especially famers constituting more than two third of the population. Having set ultimate goal of benefitting society at large, the university intends to achieve it by imparting word-class need based agricultural education, research, extension and public service.

Browse

20161
Reset filters
Subject: Statistics × Author: Rai, Gaurav Kumar × Start date: 2010 ×
Reset filters

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    ThesisItemOpen Access
    Construction of Efficient Two-level Supersaturated Designs
    (Department of Statistics, Mathematics and Computer Application, BAU, Sabour, 2016-08) Rai, Gaurav Kumar; Kole, Basudev
    Supersaturated designs are factorial designs in which the number of runs (row) is less than the total number of factors (column). They are commonly used in screening experiments, with the objective of identifying screening out the important factors from a large set of potentially active variables. The huge advantage of these designs is that they reduce the experimental cost drastically, but their critical disadvantage is the confounding involved in the statistical analysis. In this thesis work we developed an algorithm for balanced and as well as nearly balanced efficient two levelsupersaturated design, with the help of the algorithms proposed by Nguyen (1996), Lejeune (2003), Ryan and Bulutoglu (2007), Gupta et al. (2008) and Gupta et al. (2010) have been modified to construct efficient SSDs for both balanced as well as nearly balanced cases. The proposed algorithm have been constructed in such a way that it generates an SSD where all the columns are distinct and no column can be generated from any other column. The algorithm generates balanced SSDs for even n and nearly balanced SSDs for odd n. The algorithm checks the efficiency of the constructed balanced design by using lower bound to E(s2), given by Das et al. (2008) and nearly balanced design by using the lower bound to E(s2) given by Suen and Das (2010). For balanced SSDs, the algorithm has been so constructed that the generated design has less number of orthogonal pair of columns. And for nearly balanced design the constructed design is efficient for all E(s2), rmax and fmax criterion.