• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.395-407
• /
• 2020

Most classification accuracy measures for optimal threshold are divided into two types: one is expressed with cumulative distribution functions and probability density functions, the other is based on ROC curve and AUC. Unal (2017) proposed the index of union (IU) as an accuracy measure that considers two types to get them. In this study, ten kinds of accuracy measures (including IU) are divided into six categories, and the advantages of the IU are studied by comparing the measures belonging to each category. The optimal thresholds of these measures are obtained by setting various normal mixture distributions; subsequently, the first and second type of errors as well as the error sums corresponding to each threshold are calculated. The properties and characteristics of the IU statistic are explored by comparing the discriminative power of other accuracy measures based on error values.The values of the first type error and error sum of IU statistic converge to those of the best accuracy measures of the second category as the mean difference between the two distributions increases. Therefore, IU could be an accuracy measure to evaluate the discriminant power of a model.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.27 no.4
• /
• pp.274-279
• /
• 2015

Wave data acquired over eleven years near Sokcho Harbor located in the central area of the east coast were analyzed using spectral method and wave-by-wave analysis method and its major wave characteristics were examined. Significant wave heights were found to be high in winter and low in summer, and peak periods were also found to be long in winter and short in summer. The maximum significant wave height observed was 8.95 m caused by the East Sea twister. The distributional pattern of the significant wave heights and peak periods were both fitted better by Kernel distribution function than by Generalized Gamma distribution function and Generalized Extreme Value distribution function. The wave data were compiled to subdivide the wave height into intervals for each month, and the cumulative occurrence rates of wave heights were calculated to be utilized for the design and construction works in nearby construction works.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.27 no.2
• /
• pp.142-147
• /
• 2015

Wave data acquired over seven years near Daejin Harbor located in the north central area of the east coast was analyzed using spectral method and wave-by-wave analysis method and its major wave characteristics were examined. Significant wave heights were found to be high in winter and low in summer, and peak periods were also found to be long in winter and short in summer. The maximum significant wave height observed was 6.59 m and was caused by Typhoon No. 1216, SANBA. The distributional pattern of the significant wave heights and peak periods were both reproduced better by Kernel distribution function than by Generalized Gamma distribution function and Generalized Extreme Value distribution function. Meanwhile, the wave data was subdivided by month and wave height level and the cumulative appearance rate was proposed to aid designing and constructing works in nearby coastal areas.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2007.12a
• /
• pp.81-83
• /
• 2007

항만이나 항로에서의 심층적인 선박운항 안전성 평가를 위한 목적으로 주로 선박운항 시뮬레이션 시스템이 사용되고 있다. 하지만, 실제해상에서 선박이 조우할 수 있는 환경 조건은 매우 다양한 반면, 비용이나 시간적인 제약으로 인하여 실시간 선박운항 시뮬레이션은 극히 한정 된 경우에 대해서만 수행되는 것이 일반적이다. 본 논문에서는, 이러한 실시간 시뮬레이션 실험 조건을 효과적이고 체계적으로 도출하기 위한 통계적 기법에 대하여 제안하고, 이 기법을 실제 선박 운항 안전성 평가를 위한 시뮬레이션 연구에 적용한 실증 분석 결과를 사례 연구로 기술하였다. 실증 분석에는 주성분을 이용한 종합 운항 난이도 산정 방법과 누적 확률분포 개념을 이용하여 선박 운항 난이도가 높은 실험 조건을 실시간 시뮬레이션 실험 조건으로 선택하는 기법을 제시하였다.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.810-811
• /
• 2007

기온의 변화에 따라 전력수요는 상관성을 가지고 밀접하게 움직이는데, 최근에 와서 상관성이 약화되는 경향을 보이면서 불확실성이 증가되고 있다. 이러한 현상이 나타나는 원인은 기온을 분포함수화하여 접근하지 못하고, 누적기온 영향력과 실질적 기온분포를 반영하지 못했기 때문이다. 따라서 이를 보완하기 위해 전력기온지수라는 개념을 새로 창출하였으며, 여기에 누적기온반응도, 유효기온분포 및 실질기온 효과를 반영하였다. 따라서 종합적 전력기상요인인 전력기온지수를 통해 전력소비자의 자발적 수요관리 유도 및 안정적 전력수급의 토대를 마련할 수 있다는 결론을 얻었다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.3
• /
• pp.1007-1015
• /
• 2013

A stochastic probability model based on the non-homogeneous Poisson process is represented that can correctly analyze the time-dependent linear and nonlinear behaviors of total damage over the occurrence process of loads. Introducing several types of damage intensity functions, the probability of failure and the total damage with respect to mean time to failure has been investigated in detail. Taking particularly the limit state to be the random variables followed with a distribution function, the uncertainty of that would be taken into consideration in this paper. In addition, the stochastic probability model has been straightforwardly applied to the rubble-mound breakwaters with the definition of damage level about the erosion of armor units. The probability of failure and the nonlinear total damage with respect to mean time to failure has been analyzed with the damage intensity functions for armor units estimated by fitting the expected total damage to the experimental datum. Based on the present results from the stochastic probability model, the preventive management for the armor units of the rubble-mound breakwaters would be suggested to make a decision on the repairing time and the minimum amounts repaired quantitatively.

• The Korean Journal of Applied Statistics
• /
• v.4 no.2
• /
• pp.157-170
• /
• 1991

This article is concerned with the goodness - of - fit test for exponentiality when both the scale and location parameters are unknown. A test procedure based on the $L_1$-norm of discrepancy between the cumulative distribution function and the empirical distribution function is proposed, and the critical values of the test statistic are obtained by Monte Carlo simulations. Also the null distributions of the proposed test statistic are presented for small sample sizes. The power of tests under certain alternative distributions is investigated to compare the proposed test statistic with the well-known EDF test statistics. Our Monte Carlo power studies reveal that the proposed test statistic has good power properties, for moderate-to-large sample sizes, in comparison to other statistics although it is a conservative test.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.36 no.12
• /
• pp.1186-1192
• /
• 2008

The reliability of gas turbine engine blades was studied. Yield strength, Young’s modulus, engine speed and gas temperature were considered as statistically independent random variables. The failure probability was calculated using five different methods. Advanced Mean Value Method was the most efficient without significant loss in accuracy. When random variables were assumed to have normal, lognormal and Weibull distributions with the same means and standard deviations, the CDF of limit state equation did not change significantly with the distribution functions of random variables. The normalized sensitivity of failure probability with respect to standard deviations of random variables was the largest with gas temperature. The effect of means and standard deviations of random variables was studied. The increase in the mean of gas temperature and the standard deviation of engine speed increased the failure probability the most significantly.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.207-207
• /
• 2023

데이터 기반 강수 예측 모델은 극한 강수 이벤트의 크기를 과소 추정하는 경향이 있다. 이는 훈련 데이터에 극한 강수 이벤트보다 일반적인 강수 이벤트가 많이 포함되어 있기 때문이다. 본 연구는 이러한 딥러닝의 데이터 불균형 문제를 해소하고자 모델을 학습시킬 때 격자별 극한 강수에 더 큰 가중치를 주어 극한 강수 예측의 정확성을 높이는 방법을 제안한다. 딥러닝 모델 중 공간-시간 필드를 정확하게 예측할 수 있는 ConvLSTM 기반 강수 예측 모델을 활용하여 레이더 강수량을 예측하였다. 먼저, 훈련 기간 동안의 강수 이벤트의 누적 분포 함수 CDF(Cummulative distribution funcion)을 그린 후 극한 강수 이벤트와 일반적인 강수 이벤트의 분포를 확인하였다. 그다음, 적은 분포를 가진 극한 강수 이벤트의 더 큰 가중치를 두어 모델을 학습시켰다. 이 모델은 대한민국 중부 지역 (200km x 200km)의 5km-10분 해상도 레이더-계량기 복합 강수 필드에 대해 2009-2014년 기간 동안 훈련 되었고 2015-2016년 동안 모델의 훈련을 검증 하였고, 2017-2018년 동안 테스트 되었다. 다양한 가중치 함수를 기반으로 훈련 시킨 결과 최적화 가중치 함수 모델의 평균 NSE는 0.6 평균 RMSE는 0.00015 그리고 극한 강수 이벤트만 따로 추출한 평균 MAE는 6이다. 결과적으로 제안된 모델은 기존 방법에 비해 예측 성능을 향상 시켰으며, 격자별 가중치를 두었을 경우 일반적인 강수 이벤트 뿐만 아니라 극한 강수 이벤트의 예측의 정확도를 향상시켰다.

• Journal of Wetlands Research
• /
• v.17 no.3
• /
• pp.251-263
• /
• 2015

This study suggested a novel approach of estimating the optimal probability density function (OPDF) of the annual maximum rainfall time series (AMRT) combining the L-moment ratio diagram and the geographical information system. This study also reported several interesting geographical characteristics of the AMRT in Korea. To achieve this purpose, this study determined the OPDF of the AMRT with the duration of 1-, 3-, 6-, 12-, and 24-hours using the method of L-moment ratio diagram for each of the 67 rain gages in Korea. Then, a map with the Thiessen polygons of the 67 rain gages colored differently according the different type of the OPDF, was produced to analyze the spatial trend of the OPDF. In addition, this study produced the color maps which show the fitness of a given probability density function to represent the AMRT. The study found that (1) both L-skewness and L-kurtosis of the AMRT have clear geographical trends, which means that the extreme rainfall events are highly influenced by geography; (2) the impact of the altitude on these two rainfall statistics is greater for the mountaneous region than for the non-mountaneous region. In the mountaneous region, the areas with higher altitude are more likely to experience the less-frequent and strong rainfall events than the areas with lower altitude; (3) The most representative OPDFs of Korea except for the Southern edge are Generalized Extreme Value distribution and the Generalized Logistic distribution. The AMRT of southern edge of Korea was best represented by the Generalized Pareto distribution.