Author:
Shabbir Javid,Hassan Jamshaid UL,Hassan Mahmood UL,Noreen Khadija,Hussain Sajid,Ahmed Rashid
Abstract
Minimal neighbor balanced designs are economical, therefore, these are preferred by the experimenters to minimize the bias due to neighbor effects. Minimal circular balanced neighbor designs cannot be constructed for almost every case of v even, where v is number of the treatments to be compared. For v even, the circular GN2-designs in which each treatment appears exactly once as neighbors with all other treatments except the one with which it appears twice, are considered the better alternate to the minimal balanced neighbor designs. In this article, an algorithm is developed to generate the circular GN2-designs for v even which can be converted directly into minimal circular balanced and strongly balanced neighbor designs. This algorithm is also coded with R.
Subject
General Earth and Planetary Sciences,General Engineering,General Environmental Science
Reference27 articles.
1. Ahmed, R., & Akhtar, M. (2011a). Designs balanced for neighbor effects in circular blocks of size six. Journal of Statistical Planning and Inference, 141, 687–691.
2. Ahmed, R., & Akhtar, M. (2011b). Construction of neighbor balanced designs in linear blocks. Communications in Statistics-Theory and Methods, 40, 1–8.
3. Ahmed, R., Akhtar, M., & Tahir, M. H. (2009). Economical generalized neighbor designs of use in serology. Computational Statistics and Data Analysis, 53, 4584–4589.
4. Akhtar, M., Ahmed, R., & Yasmin, F. (2010). A catalogue of nearest neighbor balanced-designs in circular blocks of size five. Pakistan Journal of Statistics, 20(2), 397–405.
5. Azais, J. M. (1987). Design of experiments for studying intergenotypic competition. Journal of the Royal Statistical Society: Series B, 49, 334–345.