• 대한안전경영과학회지
• /
• 제1권1호
• /
• pp.91-100
• /
• 1999

We propose a decision model to incorporates the values assigned by a group of experts on different factors in selecting robots. Using this model, SN ratio of taguchi method for each of subjective factors as well as values of weights are used in this comprehensive method for robot selection. A numerical example is presented to illustrate the model and to show a rank reversal when compared to a model that does not eliminate extreme values and eliminates the highest and lowest experts' values allocating the weights and the subjective factors.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 1999년도 추계학술대회
• /
• pp.537-540
• /
• 1999

We propose a decision model to incorporates the values assigned by a group of experts on different factors in selecting robots. Using this model, SN ratio of taguchi method for each of subjective factors as well as values of weights are used in this comprehensive method for robot selection. A numerical example is presented to illustrate the model and to show a rank reversal when compared to a model that does not eliminate extreme values and eliminates the highest and lowest experts' values allocating the weights and the subjective factors.

• Communications for Statistical Applications and Methods
• /
• 제13권2호
• /
• pp.309-318
• /
• 2006

There are many statistical methods of testing the equality of two population variances. Among them, the well-known F test is very sensitive to the normality assumption. Several other tests that do not assume normality have been proposed, but these tests usually need tables of critical values or software for hypotheses testing. McGrath and Yeh (2005) suggested a quick and compact Count Five test requiring only the calculation of the number of extreme points. Since the Count Five test uses only extreme values, this discards some information from the samples, often resulting in a degradation in power. In this paper, an alternative Count Five test using the trimmed mean is proposed and its properties are discussed for some distributions and normal mixtures.

• Wind and Structures
• /
• 제16권4호
• /
• pp.355-372
• /
• 2013

The Gringorten estimator has been extensively used in extreme value analysis of wind speed records to obtain unbiased estimates of design wind speeds. This paper reviews the derivation of the Gringorten estimator for the mean plotting position of extremes drawn from parents of the exponential type and demonstrates how it eliminates most of the bias caused by the classical Weibull estimator. It is shown that the coefficients in the Gringorten estimator are the asymptotic values for infinite sample sizes, whereas the estimator is most often used for small sample sizes. The principles used by Gringorten are used to derive a new Consistent Linear Unbiased Estimator (CLUE) for the mean plotting positions for the Fisher Tippett Type 1, Exponential and Weibull distributions and for the associated standard deviations. Analytical and Bootstrap methods are used to calibrate the bias error in each of the estimators and to show that the CLUE are accurate to better than 1%.

• Communications for Statistical Applications and Methods
• /
• 제31권1호
• /
• pp.37-53
• /
• 2024

In this study, flood records from 79 sites across Thailand were analyzed to estimate flood indices using the regional frequency analysis based on the L-moments method. Observation sites were grouped into homogeneous regions using k-means and Ward's clustering techniques. Among various distributions evaluated, the generalized extreme value distribution emerged as the most appropriate for certain regions. Regional growth curves were subsequently established for each delineated region. Furthermore, 20- and 100-year return values were derived to illustrate the recurrence intervals of maximum rainfall across Thailand. The predicted return values tend to increase at each site, which is associated with growth curves that could describe an increasing long-term predictive pattern. The findings of this study hold significant implications for water management strategies and the design of flood mitigation structures in the country.

• 응용통계연구
• /
• 제24권5호
• /
• pp.833-845
• /
• 2011

본 논문에서는 1998.01.03부터 2011.08.31까지 수집된 코스피 지수 자료로부터 계산된 일별 로그수익률과 일별 로그손실률에 대한 극단값 통계분석을 수행하였다. 사용된 극단값 통계분석 모형은 포아송-GPD 모형이고 모수의 추정과 극단분위수의 추정은 최대가능도 방법을 적용하였다. 본 논문에서는 또한 포아송-GPD 모형에 추가적으로 모수의 무정보사전분포를 가정한 베이지안 방법을 고려하였다. 여기서는 마르코프 연쇄 몬테칼로 방법을 적용하여 모수와 극단분위수를 추정하였다. 분석 결과 최대가능도 방법과 베이지안 방법에서 모두, 로그수익률 분포의 오른쪽 꼬리는 정규분포보다 짧은 반면, 로그손실률 분포의 오른쪽 꼬리는 정규분포보다 두텁다는 결론이 얻어졌다. 극단값 분석에서 베이지안 방법을 사용할 때의 장점은 정칙조건이 만족되지 않는 경우에도 최대가능도추정량의 전통적 점근 성질을 걱정할 필요가 없고 예측의 경우에는 모수의 불확실성과 미래 관측치의 불확실성이 모두 반영되는 효과가 있다는 것이다.

• 응용통계연구
• /
• 제23권5호
• /
• pp.835-844
• /
• 2010

극단값 통계 분석의 도구로는 전통적인 연간 최대값 방법과 현대적인 분계점 방법, 그리고 분계점 방법을 개선한 변형체 등으로 분류할 수 있다. 연간 최대값 방법은 시계열자료의 연간 최대값들에 대하여 일반화극단값분포를 적합시키는 것이고, 분계점 방법은 충분히 큰 하나의 분계점을 넘어서는 초과값들의 초과여분들에 대하여 일반화파레토분포를 적합시키는 것이다. 분계점 방법의 한 변형체로서 본 논문에서는 분계점 방법에 추가적으로 초과값들의 전체 개수가 포아송분포를 따른다고 가정하는 포아송-GPD 방법을 다루고, 이를 1988.01.04부터 2009.12.31까지 수집된 서부텍사스산중질유의 현물가격 자료로부터 계산된 일일 상승율과 일일 하락율에 적용한다. 이에 따르면 일일 상승율과 일일 하락율의 분포는 정규분포와 달리 두터운 꼬리를 갖는 분포로 나타났는데, 이는 오늘날의 많은 금융 자료분석에서 나타나는 일반적인 현상과 잘 부합하는 것이다.

• Journal of the Korean Data and Information Science Society
• /
• 제27권5호
• /
• pp.1155-1168
• /
• 2016

기상 자료의 경우 한 지역의 기후가 인접지역의 기후와 비슷한 양상을 띄고 각 지역의 확률 밀도 함수 (probability density function)가 잘 알려진 확률 모형을 따르지 않는다는 것이 알려져 있다. 본 논문에서는 이러한 특성을 고려하여 이상 기후 현상이 뚜렷히 나타나는 여름철 평균 극한 기온(extreme temperature)의 확률 밀도 함수를 추정하고자 한다. 이를 위하여 공간적 상관관계 (spatial correlation)를 고려하는 비모수 베이지안 (nonparametric Bayesian) 모형인 조건부 자기회귀 종추출 혼합모형 (mixtures of conditional autoregression species sampling model)을 이용하였다. 자료는 이스트앵글리아 대학교 (University of East Anglia)에서 제공하는 전 지구의 최대 기온과 최소 기온자료 중 우리나라에 해당하는 지역의 자료를 사용하였다.

• 한국태양에너지학회 논문집
• /
• 제32권1호
• /
• pp.15-24
• /
• 2012

IEC 61400-1 requires design lifetime of wind turbines at least 20 years, thus wind turbine should be assured for structural safety through load assessment. DLCs have been defined with respect to the load assessment in IEC 61400-1. In addition, if the extreme design values for DLC1.3 are equal or exceed the extreme design value for DLC1.1, DLC1.1 may be omitted. To omit DLC1.1, scale factor (c) will be increased in DLC1.3. However, this particular adjustment is not specified guidelines. Thus, this study was conducted. DLC1.1 was calculated for extrapolation of 50 years-extreme events using several probability distribution functions and fitting methods. And DLC1.3 was calculated for up to seven different values of scale factor (c) with $2{\leq}c{\leq}5$ in steps of 0.5. Finally, in this study, scale factor (c) that was the value of 4.51 was determined.

• Journal of Electrical Engineering and Technology
• /
• 제11권4호
• /
• pp.993-1002
• /
• 2016

Hepatitis is a major public health problem all around the world. This paper proposes an automatic disease diagnosis system for hepatitis based on Genetic Algorithm (GA) Wavelet Kernel (WK) Extreme Learning Machines (ELM). The classifier used in this paper is single layer neural network (SLNN) and it is trained by ELM learning method. The hepatitis disease datasets are obtained from UCI machine learning database. In Wavelet Kernel Extreme Learning Machine (WK-ELM) structure, there are three adjustable parameters of wavelet kernel. These parameters and the numbers of hidden neurons play a major role in the performance of ELM. Therefore, values of these parameters and numbers of hidden neurons should be tuned carefully based on the solved problem. In this study, the optimum values of these parameters and the numbers of hidden neurons of ELM were obtained by using Genetic Algorithm (GA). The performance of proposed GA-WK-ELM method is evaluated using statical methods such as classification accuracy, sensitivity and specivity analysis and ROC curves. The results of the proposed GA-WK-ELM method are compared with the results of the previous hepatitis disease studies using same database as well as different database. When previous studies are investigated, it is clearly seen that the high classification accuracies have been obtained in case of reducing the feature vector to low dimension. However, proposed GA-WK-ELM method gives satisfactory results without reducing the feature vector. The calculated highest classification accuracy of proposed GA-WK-ELM method is found as 96.642 %.