• 응용통계연구
• /
• 제11권1호
• /
• pp.53-64
• /
• 1998

수리 가능한 시스템에 대한 고장시간을 표현하는 여러가지 형태의 통계적 모형이 최근 엔지니어들과 신뢰성분야 학자들의 많은 관심을 끌고 있다. 본 논문에서는 수리가능한 시스템의 신뢰성 증가를 나타내는데 유용하게 적용되는 power law process를 고려하고 특히 정시중단자료(time truncated data)인 경우 고장간격에 대한 신뢰구간을 붓스트랩 기법을 이용하여 구하고 이것을 Crow(1982)가 구한 기존의 신뢰구간과 비교 분석하였다.

• Communications for Statistical Applications and Methods
• /
• 제19권1호
• /
• pp.147-155
• /
• 2012

We propose some properties for a fuzzy correlation test by reduced-spread fuzzy variance for sample fuzzy data. First, we define the condition of fuzzy data for repeatedly observed data or that which includes error term data. By using the average of spreads for fuzzy numbers, we reduce the spread of fuzzy variance and define the agreement index for the degree of acceptance and rejection. Given a non-normal random fuzzy sample, we have bivariate normal distribution by apply Box-Cox power fuzzy transformation and test the fuzzy correlation for independence between the variables provided by the agreement index.

• 응용통계연구
• /
• 제15권2호
• /
• pp.415-421
• /
• 2002

통계적분석에서 가장 대표적인 가정이 정규성 가정이므로 데이터의 정규성 검정은 매우 중요하다. 이 논문에서는 정규성 검정을 위해 경제학에서 소득분배의 불균형에 관한 척도로 널리 이용되는 Lorenz curve를 변형한 새로운 플롯과 검정통계량을 제시한다. 그리고 제한한 검정을 W검정 (Shapiro and Wilk (1965)), Lorenz curve를 이용한 TL검정(Kang and Cho (1999))과 몬테칼로 방법을 이용하여 검정력을 비교한다. 제안된 검정이 특별한 대립분포의 경우를 제외하고는 대부분 검정력이 높았다.

• 한국전기전자재료학회논문지
• /
• 제17권8호
• /
• pp.898-903
• /
• 2004

In this paper, neural networks is studied to apply as a PD source classification in XLPE power cable specimen. For treeing discharge sources in the specimen, three defected models are made. And these data making use of a computer-aided discharge analyser, statistical and other discharge parameters is calculated to discrimination between different models of discharge sources. And also these parameter is applied to classify PD sources by neural networks. Neural Networks has good recognition rate for three PD sources.

• Journal of the Korean Statistical Society
• /
• 제33권1호
• /
• pp.129-147
• /
• 2004

For square contingency tables with nominal categories, this paper proposes a measure to represent the degree of departure from the quasi-symmetry (QS) model and the Bradley-Terry (BT) model. The measure proposed is expressed by using the Cressie and Read (1984)'s power-divergence or Patil and Taillie (1982)'s diversity index. The measure lies between 0 and 1, and it is useful for comparing the degree of departure from QS or BT in several tables.

• 대한전자공학회논문지
• /
• 제17권2호
• /
• pp.15-22
• /
• 1980

Graphic image data의 효율적인 coding을 위해서, 한글 image와 영문 image의 통계적인 특성들이 측정 비교되었다. 또한 Markov model에 의한 run length의 확률분포와 측정된 run length의 확률분포가 비교 검토되었다. 측정된 run length 분포는 negative-power 분포에 유사하며, 이것은 한글 image 에서 더욱 뚜렷한 것으로 나타났다. 대표적 네가지 run length code의 coding 특성이 비교되었는데, 영문 image보다 한글image의 coding 특성이 더욱 우수한 것으로 밝혀졌다.

• Wind and Structures
• /
• 제31권5호
• /
• pp.419-430
• /
• 2020

Weibull distribution is a conspicuous distribution known for its accuracy and its usage for wind energy analysis. The two and three parameter Weibull distributions are adopted in this study to fit wind speed data. The daily mean wind speed data of Ennore, Tamil Nadu, India has been used to validate the procedure. The parameters are estimated using maximum likelihood method, least square method and moment method. Four statistical tests namely Root mean square error, R2 test, Kolmogorov-Smirnov test and Anderson-Darling test are employed to inspect the fitness of Weibull probability density functions. The value of shape factor, scale factor, wind speed and wind power are determined at a height of 100m using extrapolation of numerical equations. Also, the value of capacity factor is calculated mathematically. This study provides a way to evaluate feasible locations for wind energy assessment, which can be used at any windy site throughout the world.

• 한국경영과학회지
• /
• 제18권2호
• /
• pp.57-81
• /
• 1993

This paper is concerned with analyzing the bankruptcy prediction power of three methods : Multivariate Discriminant Analysis (MDA), Inductive Learning, Neural Network, MDA has been famous for its effectiveness for predicting bankrupcy in accounting fields. However, it requires rigorous statistical assumptions, so that violating one of the assumptions may result in biased outputs. In this respect, we alternatively propose the use of two AI models for bankrupcy prediction-inductive learning and neural network. To compare the performance of those two AI models with that of MDA, we have performed massive experiments with a number of Korean bankrupt-cases. Experimental results show that AI models proposed in this study can yield more robust and generalizing bankrupcy prediction than the conventional MDA can do.

• Journal of the Korean Statistical Society
• /
• 제30권4호
• /
• pp.679-691
• /
• 2001

Two truncated sequential tests are developed for the two-sample scale problem based on the usual Wilcoxon rank-sum statistic for two different dispersion indices - absolute median deviations, when the medians of the two populations X and Y are equal or known and sums of squared mean deviations, when the medians are either unknown or unequal. The first test is briefly called SWAMD test and the second SWSMD test. For the SWAMD test, the percentile points for both the one-sided and two-sided alternatives, (equation omitted) have been found by Wiener approximation and their values computed for a range of values of a and N; analytical expression for the power function has been derived through Wiener process and its performance studied for various sequential designs for exponential distribution. This test has been illustrated by a numerical example. All the results of the SWAMD test, being directly applicable to the SWSMD test, are not dealt with separately Both the tests are compared and their suitable applications indicated.

• Journal of the Korean Statistical Society
• /
• 제26권4호
• /
• pp.507-520
• /
• 1997

In this paper we propose a robust Wald-type test which is based on an efficient Mallows-type one-step GM-estimator. The proposed estimator based on the weight function of Song, Park and Nam (1996) has a bounded influence function and a high breakdown point. Under some regularity conditions, we compute the finite-sample breakdown point, and drive asymptotic normality of the proposed estimator. The level and power breakdown points, influence function and asymptotic distribution of the proposed test statistic are main points of this paper. To compare the performance of the proposed test with other tests, we perform some Monte Carlo simulations.