Construction of Balanced n-ary Designs via Complete Multiset Spaces

Area of Science:

• Statistics
• Combinatorics
• Experimental Design

Background:

• Block designs are essential for efficient experimentation.
• Existing designs often lack flexibility for repeated treatments within blocks.
• Applications in dose-response and repeated-measures experiments require specialized designs.

Purpose of the Study:

• To develop balanced n-ary block designs accommodating repeated treatments.
• To establish a theoretical framework linking combinatorial symmetry to statistical optimality.
• To provide explicit formulas for design parameters and performance metrics.

Main Methods:

• Introduction of a master (k+1)-ary design with replication and concurrence parameters.
• Symmetry-preserving deletions to generate infinite families of lower-arity designs.
• Derivation of formulas for replication numbers, pairwise concurrences, and information matrices.
• Simulation studies to compare proposed designs with random arrangements.

Main Results:

• Infinite families of balanced n-ary block designs with repeated treatments were generated.
• Explicit formulas for design parameters and statistical performance were obtained.
• The proposed multiset-based designs demonstrated lower contrast variances due to equireplication.
• A direct link between combinatorial Sv-symmetry and statistical optimality was established.

Conclusions:

• The developed designs provide a rigorous and statistically optimal approach for experiments with repeated treatments.
• These designs are advantageous over random multiset arrangements, especially in dose-response and repeated-measures studies.
• The findings offer a valuable tool for enhancing the efficiency and reliability of experimental designs.