The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Shift-Power Transformation

이동-멱변환에 관한 연구

  • Cho Ki-Jong (Division of Environmental Science and Ecological Engineering, Korea University) ;
  • Jeong Seok-Oh (Department of Statistics, Hankuk University of Foreign Studies) ;
  • Shin Key-Il (Department of Statistics, Hankuk University of Foreign Studies)
  • 조기종 (고려대학교 환경생태학부) ;
  • 정석오 (한국외국어대학교 정보통계학과) ;
  • 신기일 (한국외국어대학교 정보통계학과)
  • Published : 2006.07.01
Download PDF
⟨ Previous Next ⟩

Abstract

Generally speaking, power transformations such as Box-Cox transformation(1964) is applied for variance stabilization and symmetry. But, when the distribution of the original data has a large mean with a small variance or the coefficient of variation is very small, they don't work at all. This paper propose a simple method to introduce a shift parameter before applying power transformations and showed the numerical evidence by Monte Carlo simulation and a real data analysis.

일반적으로 Box-Cox변환과 같은 류의 멱변환은 분산 안정화 혹은 분포의 대칭성 향상 등을 목적으로 사용된다. 그러나 원 자료의 평균의 크기가 크면서 분산이 상대 적으로 작은 경우, 즉 변동계수가 작은 경우에는 제대로 작동하지 않는 것이 알려져 있다. 본 논문에서는 이러한 문제점을 해결하기 위한 이동-멱변환을 제안하고 모의실험과 실제 자료 분석을 통하여 그 효과를 확인하였다.

Keywords

References

  1. Atkinson, A. C., Pericchi, L. R. and Smith, R. L. (1991). Grouped likelihood for the shifted power transformation, Journal of the Royal Statistical Society B 53, 473-482
  2. Bickel, P. J. and Doksum, K. A. (1981). An analysis of transformations revisited, Journal of the American Statistical Association 76, 296-311 https://doi.org/10.2307/2287831
  3. Bloch, D. A. and Gastwirth, J. L. (1968). On a simple estimate of the reciprocal of the density function, Annals of Mathematical Statistics 39, 1083-1085 https://doi.org/10.1214/aoms/1177698342
  4. Box, G. E. P. and Cox, D. R. (1964). An analysis of transformations (with Discussion), Journal of the Royal Statistical Society B 26, 211-252
  5. Cochran, W. G. (1977), Sampling Techniques. 3rd edition, Wiley
  6. Hernandez, F. and Johnson, R. A. (1980). The large-sample behavior of transformations to normality, Journal of the American Statistical Association 75, 855-861 https://doi.org/10.2307/2287172
  7. Park, H. and Shin, K.-I. (2006). A shrinked forecast in stationary processes favouring percentage error, Journal of Time Series Analysis 27, 129-139 https://doi.org/10.1111/j.1467-9892.2005.00458.x
  8. Shin, K.-I. and Kang, H. (2001). A study of the effect of power transformation in the ARMA(p,q) model, Journal of Applied Statistics 28, 1019-1028 https://doi.org/10.1080/02664760120076689
  9. Yeo, I.-K. and Johnson, R. A. (2000). A new family of power transformations to improve normality or symmetry, Biometrika 87, 954-959 https://doi.org/10.1093/biomet/87.4.954