Bhartiya Krishi Anusandhan Patrika
  • Year: 2022
  • Volume: 37
  • Issue: 1

Partially balanced 3-designs using mutually orthogonal latin squares

  • Author:
  • Sayantani Karmakar1, Cini Varghese1, Seema Jaggi2, Mohd Harun1,*, Devendra Kumar1
  • Total Page Count: 6
  • Page Number: 7 to 12

1ICAR-Indian Agricultural Statistics Research Institute, Pusa-110 012, New Delhi, India

2Human Resource Development, ICAR, New Delhi-110 001, India

*Corresponding Author: Mohd Harun, ICAR-Indian Agricultural Statistics Research Institute, Pusa-110 012, New Delhi, India. Email: mohd.harun@icar.gov.in

Online published on 26 July, 2022.

Abstract

t-designs represent a generalized class of balanced incomplete block designs in which the number of blocks in which any t treatments (t>2) occur together is a constant. Like other families of incomplete block designs, t-designs find potential application in farming system research where the main concern would be to select the best combination out of a certain set of t-component farming systems for a specific agro-ecological zone. But in order to obtain t-balance we may require large number of blocks and replications. Partially balanced t-designs are introduced for experimental situations where it is not possible to get t-balanced designs. A series of partially balanced 3-designs are constructed using mutually orthogonal latin squares. These designs, though possess pair-wise balance, are not combinatorially balanced in terms of 3-tuples. They also satisfy the properties of 3-packing designs. The efficiency factor of these designs is quite high as v increases, the efficiency increases.

Keywords

Balanced incomplete block designs, Canonical efficiency factor, Mutually orthogonal latin squares, T-packing