search for




 

Power analysis for 6×5 Youden square design
Journal of the Korean Data & Information Science Society 2018;29:1127-36
Published online September 30, 2018
© 2018 Korean Data and Information Science Society.

Young Hun Choi1

1Department of Applied Statistics, Hanshin University
Correspondence to: Professor, Department of Applied Statistics, Hanshin University, 447-791, Korea. E-mail: choicyh@hs.ac.kr
This research was supported by Hanshin University Research Grant.
Received July 18, 2018; Revised August 22, 2018; Accepted August 28, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
This research compares power between the parametric statistic and rank transformed statistic for testing treatment effect in fixed 6×5 Youden square design reflecting the adjusted sum of squares for treatment. Powers for the parametric and rank transformed statistic with asymmetric and unequally spaced two blocks and treatment effects are positioned in the top. Powers with symmetric and equally spaced effects are ranked in the middle. Powers with symmetric and unequally spaced effects are placed in the lowest. In addition powers of the rank transformed statistic are superior when block effects size are relatively smaller than treatment effect size and row block effect size is smaller than column block effect size. Meanwhile powers of the rank transformed statistic show relative preference to those of the parametric statistic for exponential and double exponential distributions, whereas powers of the parametric and rank transformed statistic maintain similar level for normal and uniform distributions.
Keywords : Parametric statistic, power, rank transformed statistic, 6×5 Youden square design.