Journal of the Korean Data and Information Science Society

The Korean Data and Information Science Society (한국데이터정보과학회)
권호 표지
DOI QR코드

DOI QR Code

Power study for 2 × 2 factorial design in 4 × 4 latin square design

4 × 4 라틴방격모형 내 2 × 2 요인모형의 검정력 연구

  • 최영훈 (한신대학교 응용통계학과)
  • Received : 2014.07.10
  • Accepted : 2014.08.21
  • Published : 2014.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

Compared with single design, powers of rank transformed statistic for testing main and interaction effects for $2{\times}2$ factorial in $4{\times}4$ latin square design are rapidly increased as effect size and replication size are increased. In general powers of rank transformed statistic are superior without regard to the diversified effect composition and the type of error distributions as nontesting factors are few and effect size are small. Powers of rank transformed statistic show much higher level than those of parametric statistic in exponential and double exponential distributions. Further powers of rank transformed statistic are very similar with those of parametric statistic in normal and uniform distributions.

반복이 존재하는 $4{\times}4$ 라틴방격모형 내 $2{\times}2$ 요인모형의 주효과 및 상호작용효과를 검정하기 위한 순위변환 통계량의 검정력은 단일모형에 비하여 효과크기 및 반복크기가 커질수록 빠르게 증가한다. 일반적으로 다양한 효과구성 및 모든 오차항 분포와 상관없이 검정하고자 하는 요인 이외의 효과가 존재하는 요인 수가 적고 효과크기가 작을수록 순위변환 통계량의 검정력은 뛰어나다. 특히 오차항이 지수분포 및 이중지수분포일 때 순위변환 통계량의 검정력은 모수적 통계량의 검정력보다 상대적으로 높은 비교우위를 보이며, 정규분포 및 균일분포에서는 전반적으로 별다른 차이가 없다. 이는 두개의 주효과, 한개의 상호작용효과 및 두개의 블럭효과 등의 다섯 가지 효과가 동시에 존재하는 다인자로 구성된 라틴방격과 요인모형의 결합형태의 특이성으로 인한 결과이다.

Keywords

References

  1. Akritas, M. G. and Papadatos, N. (2004). Heteroscedastic one way ANOVA and lack of fit tests. Journal of the American Statistical Association, 99, 368-382. https://doi.org/10.1198/016214504000000412
  2. Choi, Y. H. (1998). A study of the power of the rank transform test in a 2(3) factorial experiment. Communications in Statistics, 27, 251-266.
  3. Choi, Y. H. (2009). Power analysis for 3 ${\times}$ 3 latin square design. Journal of the Korean Data & Information Science Society, 20, 401-410.
  4. Choi, Y. H. (2011). Power analysis for 2 x 2 factorial in randomized complete block design. Journal of the Korean Data & Information Science Society, 22, 245-253.
  5. Choi, Y. H. (2012). Power study for 4 ${\times}$ 4 Graeco latin square design. Journal of the Korean Data & Information Science Society, 23, 683-691. https://doi.org/10.7465/jkdi.2012.23.4.683
  6. Hicks, C. R. (1982). Fundamental concepts in the design of experiments, 3rd edition, Holt, Rinehart and Winston, New York.
  7. Hora, S. C. and Iman, R. L. (1988). Asymptotic relative efficiencies of the rank transformation procedure in randomized complete block design. Journal of the American Statistical Association,83, 462-470. https://doi.org/10.1080/01621459.1988.10478618
  8. Jo, S. and Kim, D. (2013). Nonparametric procedures using aligned method and joint placement in randomized block design. Journal of the Korean Data & Information Science Society, 24, 95-103. https://doi.org/10.7465/jkdi.2013.24.1.95
  9. Kepner, J. L. and Robinson, D. H. (1988). Nonparametric methods for detecting treatment effects in repeated-measures designs. Journal of the American Statistical Association, 83, 456-461. https://doi.org/10.1080/01621459.1988.10478617
  10. Montgomery, D. C. (1991). Design and analysis of experiments, 3rd edition, John Wiley & Sons, New York.
  11. Neter, J., Wasserman, W. and Kutner, M. (1990). Applied linear statistical model, 3rd edition, Irwin, Homewood and Boston.
  12. Pavur, R. and Nath, R. (1986). Parametric versus rank transform procedures in the two-way factorial experiment. Journal of Statistical Computation Simulation,23, 231-240. https://doi.org/10.1080/00949658608810873
  13. Pirie, W. R. and Rauch, H. L. (1984). Simulated efficiencies of tests and estimators from general linear models analysis based on ranks: The two-way layout with interaction. Journal of Statistical Computation Simulation,20, 197-204. https://doi.org/10.1080/00949658408810776

Cited by

  1. 3×3 분할요인모형의 검정력 비교연구 vol.28, pp.1, 2014, https://doi.org/10.7465/jkdi.2017.28.1.143