Water for future (물과 미래)

Korea Water Resources Association (한국수자원학회)
권호 표지

Separation Effect of Rainfall Data Based on Parameter Estimation Methods

매개변수 추정방법에 따른 강우자료의 분리효과

  • 김경덕 (연세대학교 대학원 토목공학과) ;
  • 배덕효 (연세대학교 공과대학 산업기술연구소 선임연구원, 연세대학교 공과대학 토목공학과 조교수, 연세대학교 공과대학 토목공학과)
  • Published : 1996.02.01
Download PDF
⟨ Previous Next ⟩

Abstract

It is very important to select appropriate distributions for hydrological data in planning and designing hydraulic structures. Also, it is necessary to check whether the selected distribution reproduces the statistical characteristics of the real data. In this study, the parameters of the two- and three-parameter gamma, two- and three-parameter lognormal, Gumbel, two- and three-parameter log-Gumbel, GEV, log-Pearsonn type III, two- and three-parameter Weibull, four- and five-parameter Wakeby distributions were estimated for the rainfall data of 22 sites in Korea with 7 different durations based on the methods of moments, probability weighted moments, and maximum likelihood. And the validity conditions were checked for the estimated parameters. The separation effect for each distribution was examined throught 10,000 simulations using the estimated parameters. As results, the separation effect was the smallest: log-Pearson type III for moment method, log-Pearson type III and GEV for probability weighted moment method, and GEV for maximum likelihood method. However, it is large for the two-parameter distributions.

수공구조물의 계획과 설계에 있어서 수문자료에 대한 적정분포형을 선정하는 것은 매우 중요하며, 선정된 분포함수가 실측자료의 통계학적 특성을 잘 나타내고 있는가를 검토하는 것은 필수적인 과제이다. 본 연구에서는 전국 22개 지점, 7개 지속기간의 강우자료에 대하여 2변수 및 3변수 gamma, 2변수 및 3변수 lognormal, Gumbel, 2변수 및 3변수 log-Gumbel, GEV, log-Pearson type III, 2변수 및 3변수 Weibull, 4변수 및 5변수 Wakeby 분포를 적용하여 모멘트법, 확률가중 모멘트법, 최우도법 등으로 각 분포형의 매개변수를 추정하고, 적합성 조건을 검사하였다. 각 매개변수 추정법에 의하여 추정된 매개변수를 이용하여 10,000번 모의 발생하여 분리효과를 검토한 결과 매개변수 적합성을 고려한 경우 모멘트법에서는 log-Pearson type III 분포, 확률가중 모멘트법에서는 log-Pearson type III와 GEV 분포, 최우도법에서는 GEV 분포가 분리효과를 가장 작게 나타냈으며, 2변수 분포형의 경우 모두 분리효과가 크게 나타났다.

Keywords

References

  1. '95 한국수자원학회 학술발표회 논문집 Wakeby 분포를 이용한 강우자료의 분리효과에 대한 연구 김경덕;허준행;조원철
  2. 한국수문학회지 v.26 no.4 강우자료의 분리효과 김양수;허준행
  3. 대한토목학회 논문집 v.13 no.1 한국 대표확률강우강도식의 유도 이원환;박상덕;최성열
  4. 한국수문학회지 v.14 no.4 강수빈도 및 강우강도의 시공적 분포분석에 관한 연구 이재준;손광익;이원환;이길춘
  5. 대한토목학회논문집 v.13 no.4 대수-Gumbel 확률분포형의 매개변수 추정과 신뢰한계 유도 허준행
  6. Water Resources Research v.11 no.5 The log-Pearson type Ⅲ distribution and its application in hydrology Bobee, B.
  7. Water Resources Research v.11 no.6 Correction of bias in the estimation of the coefficient of skewness Bobee, B.;Robitaille, R.
  8. Water Resources Research v.13 no.2 The use of the Pearson type Ⅲ and log-Pearson type Ⅲ distributions revisited Bobee, B.;Robitaille, R.
  9. Statistical methods for research workers Fisher, M.
  10. Water Resources Research v.15 no.5 Probability weighted moments: Definition and relation to parameters of several distributions expressible in inverse form Greenwood, J.A.;Landwehr, J.M.;Matalas, N.C.;Wallis, J.R.
  11. Report, IBM Research Report RC12210 The theory of probability weighted moments Hosking, J.R.M.
  12. Ph.D. dissertation, Harvard University Robust estimation of use frequency of extreme events in a flood frequency context Houghton, J.C.
  13. Water Resources Research v.14 no.6 Birth of a parent: The Wakeby distribution for modeling flood flows Houghton, J.C.
  14. Water Resources Research v.14 no.6 The incomplete means estimation procedure applied to flood frequency analysis Houghton, J.C.
  15. WMO Tech. Note 98 Statistics of extremes: Estimation of maximum floods Jenkinson, A.F.
  16. Intermat. Symp. River. Mech. Proc. IAHR v.2 Some aspects for the selection of an adequate probability distribution function for flood analysis Koberg, D.;Eggers, H.
  17. Water Resources Research v.14 no.3 Some comparisons of flood statistics in real and log space Landwehr, J.M.;Matalas, N.C.;Wallis, J.R.
  18. Water Resources Research v.15 no.6 Estimation of parameters and quantiles of Wakeby distributions: 1. Known lower bounds Landwehr, J.M.;Matalas, N.C.;Wallis, J.R.
  19. Water Resources Research v.15 no.6 Estimation of parameters and quantiles of Wakeby distributions: 2. Unknown lower bounds Landwehr, J.M.;Matalas, N.C.;Wallis, J.R.
  20. Journal of Hydrology v.10 no.3 A comparison of methods of fitting the double exponential distribution Lowery, M.D.;Nash, J.E.
  21. Water Resources Research v.11 no.6 Regional skew in search of a parent Matalas, N.C.;Slack, J.R.;Wallis, J.R.
  22. Water Resources Research v.9 no.2 Eureka! It fits a Pearson type Ⅲ distribution Matalas, N.C.;Wallis, J.R.
  23. Flood studies report, Hydrological studies v.1 National Environment Research Council
  24. Water Resources Research v.20 no.7 Two-component extreme value distribution for flood frequency analysis Rossi, F.;Fiorentino, M.;Versace, P.
  25. Nord. Hydrol. v.3 no.4 Statistical interpretation of hydrometeorological extreme values Samuelson, B.
  26. Inst. fur Hydraulik und Gew sserkunde, Tech. Univ. M nchen, Mitt. Nr. 11 Verwendung parameterfreier Wahrscheinlichkeitsverteilungen zur Beschreibung hydrologischer Prozesse Siegerstetter, L.A.
  27. Mathematische Statistik in der Technik Smirnow, N.;Dunin, W.;Barkowski, I.W.
  28. Water Resources Research v.8 no.6 Fitting of distribution functions by nonlinear least squares Snyder, W.M.
  29. Mon. Weather Rev. v.86 no.4 A note on the gamma distribution Thom, H.C.S.
  30. Ann. Math. Statist. v.17 Minimal variance and its relation to efficient moment-test Vatnsdal, J.R.