라틴-하이퍼큐브 실험게획 간의 거리 계산과 비교

  • 박정수 (전남대학교 정보통신연구소 및 자연과학대하교 통계학과) ;
  • 황현식 (전남대학교 자연과학대학 통계학과)
  • 발행 : 2000.09.01

초록

전산실험계획으로 유용하게 쓰이는 라틴-하이퍼큐브 계획간의 거리를 정의하고 그 기대값을 계산하였다. 이 계산을 위해서 차원이 증가함에 따라 수리 통계학적 방법, 수치 해석적 방법(다차원 수치 적분법), 몬테카를로 적분 방법, 극한 정규분포이론을 이용하여 거리의 기대값을 구했다. 또한 같은 구조를 가지면서 랜덤성에 차이가 있는 두 라틴-하이퍼큐브 계획 간에 반응함수의 평균에서의 차이 및 정보량의 차이를 다루었다. 본 논문에서 제시한 두 Lhd들간의 비교 기법은 두 개의 일반 실험계획의 비교에도 유용하리라 여겨진다.

A distance measure between two Latin-hypercube designs is defined and its expected value is computed. It was computed by using mathematical statistics, numerical analysis (multidimensional numerical integration), Monte-carlo method, and the theory of asymptotic normal distribution. For the comparison of two Latin-hypercube designs with same structure but different randomness, the difference of expected values of response function and information mass of experimental designs are considered. These methods may be useful in comparison between two general experimental designs.

키워드

참고문헌

  1. 한국통계학회논문집 v.6 The limits of bivariate Q-Q plots based on matching that minimized a distance 김남현
  2. 응용통계연구 v.8 이단계 Latin Hypercube추출법과 그 응용 임미정;권우주;이주호
  3. Graph Theory with Applications Bondy, J. A.;Murty, U. S. R.
  4. Journal of the American Statistical Association v.85 A multivariate generalization of quantilequantile plots Easton, G. S.;McCulloch, R. E.
  5. Computers and Geosciences v.26 MINVAR and UNCCON, computer programs for uncertainty analysis of solubility calculations in geological systems Ekberg, C.;Borjesson, S.;Emren, A. T.;Samuelsson, A.
  6. Random Structure and Algorithms v.7 Perfect matchings in random s-uniform hypercube Frieze, A.;Janson, S.
  7. Environmental Engineering Science v.16 Stochastic modeling of exposure and risk in a contaminated heterogeneous aquifer. 2: Application of Latin hypercube sampling Lahkim, M. B.;Garcia, L. A.;Nuckols, J. R.
  8. Journal of the Korean Statistical Society v.28 Asymptotic comparison of Latin hypercube sampling and its stratified version Lee, J.
  9. Technometrics v.21 A comparison of three methods for selecting values of input variables in the analysis of output from a computer code McKay, M. D.;Beckman, R. J.;Conover, W. J.
  10. Ann. Statist v.22 Asymptotic Bayes criteria for nonparametric response surface design Mitchell, T.;Sacks, J.;Ylvisaker, D.
  11. Journal of Statistical Planning and Inference v.43 Exploratory designs for computational experiments Morris, M.;Mitchell, T.
  12. Journal of the Royal Statistical Society, B v.54 A central limit theorem for Latin hypercube sampling Owen, A. B.
  13. Journal of Statistical Planning and Inference v.39 Optimal Latin-hypercube designs for computer experiments Park, J. S.
  14. Water Research v.34 Risk analysis using stochastic reliability methods applied to two cases of deterministic water quality models Portielje, R.;Hvitved-Jacobsen, T.;Schaarup-Jensen, K.
  15. Stochastic Simulation Ripley, B. D.
  16. Statistical Science v.4 Designs and analysis of computer experiments Sacks, J.;Welch, W.;Mitchell, T.;Wynn, H.
  17. Technometrics v.29 Large sample properties of simulation using Latin hypercube sampling Stein, M.
  18. Annals of Applied Probability v.2 Matching random samples in many dimensions Talagrand, M.
  19. Journal of the American Statistical Association v.88 OA-Based Latin hypercubes Tang, B.