Construction of Balanced n-ary Designs via Complete Multiset Spaces
Abstract
We study balanced n-ary block designs that allow repeated treatments within blocks, motivated by applications in dose-response and repeated-measure experiments. We introduce a master (k+1)-ary design with explicit replication and pairwise concurrence parameters. Systematic symmetry-preserving deletions generate infinite families of lower-arity balanced designs, for which explicit formulas for replication numbers, structured expressions for pairwise concurrences, and the resulting information matrix under the fixed-effects model are obtained. These results establish a rigorous link between combinatorial Sv-symmetry and statistical optimality. Simulation studies show that the proposed multiset-based designs, by maintaining equireplication, yield lower contrast variances than random multiset arrangements with unequal replication numbers. Potential applications include doseresponse studies, resampling-based simulations, and repeated-measures experiments. AMS Subject Classification (2020): 62K10 · 05B05
Keywords
Citation Information
@article{sudeshsrivastav2026,
title={Construction of Balanced n-ary Designs via Complete Multiset Spaces},
author={Sudesh Srivastav and Apurv Srivastav},
journal={Research Square},
year={2026},
doi={https://doi.org/10.21203/rs.3.rs-8595082/v1}
}
SinoXiv