Abstract:This paper addresses the task to devise a statistical estimation procedure in an event where the volume of the array of initial data used in processing is insufficient to correctly determine the parameters of the response function. The object of research is the technology of statistical processing of a small sample of data. The subject of the study is the methods of statistical estimation under conditions of a small sample of initial data. The main direction is to devise a special procedure for statistical processing of a small sample of initial data, which provides a correct statistical estimation of the parameters of the response function. The method for solving the problem is the selection of the most representative orthogonal replica-like subplan from the plan of a complete factorial experiment obtained by artificially orthogonalizing the results of a passive experiment. The necessity and expediency of the proposed procedure is a consequence of the unpredictability and uneven distribution of points in the phase space of coordinates. The result of the implementation of the corresponding procedure is a truncated orthogonal plan of the full factorial experiment, which provides the possibility of independent estimation of all coefficients of the regression polynomial describing the response function. Under conditions of a severe shortage of the number of measurements, the procedure makes it possible to isolate a representative orthogonal replica from the resulting plan of a complete factorial experiment. Using this subplan of the full factorial experiment plan makes it possible to evaluate all the coefficients of the regression polynomial that describes the desired response function. The corresponding computational procedure is based on solving the triaxial Boolean assignment problem

How One-Factor-at-a-Time Experimentation Can Lead to Greater Improvements Than Orthogonal Arrays

Coarse optimization for complex systems: an application of orthogonal experiments

Optimal orthogonal designs for experiments with four-level and two-level factors

Orthogonal Array Based Locally D-Optimal Designs for Binary Responses in the Presence of Factorial Effects

Nearly Resolution V Plans on Blocks of Small Size

Identification of Single-Treatment Effects in Factorial Experiments

Causal Inference of Social Experiments Using Orthogonal Designs

A multivariate central limit theorem for randomized orthogonal array sampling designs in computer experiments

Statistical processing of a small sample of raw data using artificial orthogonalisation technology

Critical Analysis of Factorial Experiments and Orthogonal Fractions

Three-level and Mixed-Level Orthogonal Arrays for Lean Designs

Half the price, twice the gain: How to simultaneously decrease animal numbers and increase precision with good experimental design

Better than a petaflop: The power of efficient experimental design

Optimal allocation of sample size for randomization-based inference from 2 K factorial designs

Randomization Inference When N Equals One

Evaluation of orthogonal composite designs for second‐order model in presence of missing observation

Forward selection and post-selection inference in factorial designs

Simple Bayesian Algorithms for Best Arm Identification

Improving Covariate Balance in 2^K Factorial Designs via Rerandomization

Parameters for the best convergence of an optimization algorithm On-The-Fly

Randomization-based Joint Central Limit Theorem and Efficient Covariate Adjustment in Randomized Block 2K Factorial Experiments