Type: | Package |
Title: | Partially Balanced t-Designs (PBtDesigns) |
Version: | 1.0.0 |
Maintainer: | Ashutosh Dalal <ashutosh.dalal97@gmail.com> |
Description: | The t-designs represent a generalized class of balanced incomplete block designs in which the number of blocks in which any t-tuple of treatments (t >= 2) occur together is a constant. When the focus of an experiment lies in grading and selecting treatment subgroups, t-designs would be preferred over the conventional ones, as they have the additional advantage of t-tuple balance. t-designs can be advantageously used in identifying the best crop-livestock combination for a particular location in Integrated Farming Systems that will help in generating maximum profit. But as the number of components increases, the number of possible t-component combinations will also increase. Most often, combinations derived from specific components are only practically feasible, for example, in a specific locality, farmers may not be interested in keeping a pig or goat and hence combinations involving these may not be of any use in that locality. In such situations partially balanced t-designs with few selected combinations appearing in a constant number of blocks (while others not at all appearing) may be useful (Sayantani Karmakar, Cini Varghese, Seema Jaggi & Mohd Harun (2021)<doi:10.1080/03610918.2021.2008436>). Further, every location may not have the resources to form equally sized homogeneous blocks. Partially balanced t-designs with unequal block sizes (Damaraju Raghavarao & Bei Zhou (1998)<doi:10.1080/03610929808832657>. Sayantani Karmakar, Cini Varghese, Seema Jaggi & Mohd Harun (2022)." Partially Balanced t-designs with unequal block sizes") prove to be more suitable for such situations.This package generates three series of partially balanced t-designs namely Series 1, Series 2 and Series 3. Series 1 and Series 2 are designs having equal block sizes and with treatment structures 4(t + 1) and a prime number, respectively. Series 3 consists of designs with unequal block sizes and with treatment structure n(n-1)/2. This package is based on the function named PBtD() for generating partially balanced t-designs along with their parameters, information matrices, average variance factors and canonical efficiency factors. |
Imports: | MASS |
License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
Encoding: | UTF-8 |
Repository: | CRAN |
RoxygenNote: | 7.2.0 |
NeedsCompilation: | no |
Packaged: | 2023-01-18 08:32:13 UTC; Ashutosh |
Author: | Sayantani Karmakar [aut, ctb], Cini Varghese [aut, ctb], Ashutosh Dalal [aut, cre], Vinaykumar LN [aut, ctb], Seema Jaggi [aut, ctb], Mohd Harun [aut, ctb] |
Date/Publication: | 2023-01-18 11:50:10 UTC |
Partially Balanced t-Designs (PBtDesigns)
Description
This package contains functions named PBtD() for generating partially balanced t-designs along with their parameters, information matrices, average variance factors and canonical efficiency factors.
Usage
PBtD(v, Series)
Arguments
v |
Number of treatments |
Series |
Series of Partially Balanced t-Designs |
Value
Three series are given for generating of partially balanced t-designs namely Series 1, Series 2 and Series 3.
Series 1 are designs having equal block sizes and with treatment structure 4(t + 1).
Series 2 are designs having equal block sizes and with treatment as a prime number.
Series 3 consists of designs with unequal block sizes and with treatment structure n(n-1)/2.
This function generates partially balanced t-designs along with their parameters, information matrices, average variance factors and canonical efficiency factors.
References
1) Karmakar, S., Varghese, C., Jaggi, S. & Harun, M. (2021)< DOI:10.1080/03610918.2021.2008436>. " Partially Balanced t-designs ".
2) Raghavarao, D. & Zhou, B. (1998)<https://doi.org/10.1080/03610929808832657> " Universal optimality of UE 3-designs for a competing effects model ".
3) Karmakar, S., Varghese, C., Jaggi, S. & Harun, M. (2022)." Partially Balanced t-designs with unequal block sizes ".
Examples
library(PBtDesigns)
PBtD(7,2)