• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.155-165
• /
• 2018

Outliers and influential observations often distort many numerical measures for data analysis. Jang and Anderson-Cook (Quality and Reliability Engineering International, 30, 1409-1425, 2014) proposed a graphical firework plot method for exploratory analysis purpose to provide a possible visualization of the trace of the impact of the possible outlying and influential observations on the univariate/bivariate data analysis and regression. They developed 3-D plot as well as pairwise plot for the appropriate measures of interest. We use firework plots as a graphical exploratory data analysis tool to detect outliers and evaluate the impact of outliers in statistical inference.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.905-914
• /
• 2011

In the field of image processing, the removal noise contamination from the original image is essential. However, due to various reasons, the occurrence of the noise is practically impossible to prevent completely. Thus, the reduction of the noise contained in images remains important. In this study, we estimate the level of noise variance based on the measurement of the relative strength of the noise, and we propose a noise reduction algorithm that uses a sigma filter. As a result, the proposed statistical noise reduction methodology provides significantly improved results over the usual sigma filtering regardless of the level of the noise variance.

• 한국방재학회:학술대회논문집
• /
• 2011.02a
• /
• pp.205-205
• /
• 2011

본 연구에서는 한반도의 유역별 대표 기상관측 지점을 선정하여 기후변화로 인하여 미래에 나타날 수 있는 가뭄의 경향성을 분석하였다. 분석을 위한 자료는 실제 강수량 자료(1974~1999년)와 A2시나리오를 따르는 5개의 GCMs(General Circulation Model) 자료를 통계적 상세화한 강수량 자료(1974~2099년)를 이용하여 산정한 지속기간 6개월의 SPI(Standardized Precipitation Index)를 사용하였다. 분석을 위한 대표 기상관측 지점으로는 춘천, 서울, 대전, 대구, 전주, 광주, 부산 지점을 선정하였으며 GCM으로는 호주(CSIRO : MK3), 미국(GFDL : CM2_1), 독일/한국(CONS : ECHO-G), 일본(MRI : CGCM2_3_2), 영국(UKMO : HADGEM1)의 GCM을 선정하였다. 가뭄의 통계적 특성을 분석하기 위하여 Mann-Kendall 검정을 통한 경향성 분석과 Wavelet Transform 분석을 통한 주기성 분석을 하였으며 Drought Spell을 이용하여 가뭄심도별 발생빈도를 보았다. 그 결과, 경향성 분석에서는 각 GCMs의 차이를 볼 수 있었으며 CSIRO : MK3.0, GFDL : CM2_1, MIUB : ECHO-G 모델에서는 전체적으로 가뭄이 완화되고 MRI : CGCM2_3_2, UKMO : HADGEM1 모델에서는 가뭄이 심화되는 것으로 나타났다. 주기성 분석에서는 춘천, 서울에서는 낮은 주기를 대전, 대구, 전주, 광주, 부산지점에서는 다소 긴 주기를 보여주었다. Drought-spell에 의한 분석에서는 전 관측지점에서 SPI의 이론적인 확률밀도 함수값과 유사하게 나타나고 있었으며 이를 통해, 미래에는 극심한 가뭄의 빈도가 증가하고 있는 것을 예측할 수 있었다.

• Environmental and Resource Economics Review
• /
• v.14 no.1
• /
• pp.201-220
• /
• 2005

Recently, we have been witnessing new records of crude oil price hikes. One question which naturally arises would be the possibility and accuracy of forecasting crude oil prices. This study tries to answer the relative predictability of futures prices compared to the forecasts based on experts system. Using WTI crude oil spot and futures prices, this study performs simple statistical comparisons in forecasting accuracy and a formal test of differences in forecasting errors. According to statistical results, WTI crude oil futures market turns out to be equally efficient relative to EIA experts system. Consequently, WTI crude oil futures market could be utilized as a market-based tool for price forecasting and/or resource allocation for both of petroleum producers and consumers.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.10a
• /
• pp.347-349
• /
• 2014

Model verification and validation (V&V) is a current research topic to build computational models with high predictive capability by addressing the general concepts, processes and statistical techniques. The hypothesis test for validity check is one of the model validation techniques and gives a guideline to evaluate the validity of a computational model when limited experimental data only exist due to restricted test resources (e.g., time and budget). The hypothesis test for validity check mainly employ Type I error, the risk of rejecting the valid computational model, for the validity evaluation since quantification of Type II error is not feasible for model validation. However, Type II error, the risk of accepting invalid computational model, should be importantly considered for an engineered products having high risk on predicted results. This paper proposes a technique named as the response-adaptive experimental design to reduce Type II error by adaptively designing experimental conditions for the validation experiment. A tire tread block problem and a numerical example are employed to show the effectiveness of the response-adaptive experimental design for the validity evaluation.

• Journal of Korean Society of Water and Wastewater
• /
• v.24 no.5
• /
• pp.617-628
• /
• 2010

The operation performance of sewer rehabilitation projects conducted with Build-Transfer- Lease contract in Korea will be evaluated using the index of infiltration and inflow (I/I). Though I/I obtained at the fourth year should be initially evaluated based on the I/I values observed for the previous three years after the completion of sewer construction, the concrete methodology have not been proposed to rely on the so called 'performance evaluation committee'. This study suggests two statistical methodology to evaluate the I/I performance; the confidence interval method and the hypothesis-testing method. Assumed ten I/I values in each year for 20 years are used in this study. Two cases are analyzed and compared; case I to use as control data all I/I values for all years obtained before the evaluation year and case II to use I/I values for only 3 years before the evaluation year. As a result, case II tends to have relatively higher scores than case I, reflecting the low mean I/I values at the initial years.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.3
• /
• pp.457-465
• /
• 2012

The purpose of this study is to analyze the association between indicators and to find statistical models based on important indicators at 'College Notifier' in Korea Council for University Education. First, Pearson correlation coefficients are used to find statistically significant correlations. By variable selection method, the important indicators are selected and their coefficients are estimated. As variable selection method, backward and stepwise methods are employed.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.2
• /
• pp.169-174
• /
• 2018

In general, the methods of the analysis of variance(ANOVA) for the continuous data and the chi-square test for the discrete data are used for statistical analysis of the effect and the association. In multidimensional data, analysis of hierarchical structure is required and statistical linear model is adopted. The structure of the linear model requires the normality of the data. A multidimensional categorical data analysis methods are used for causal relations, interactions, and correlation analysis. In this paper, Bayesian network model using probability distribution is proposed to reduce analysis procedure and analyze interactions and causal relationships in categorical data analysis.

• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.3-8
• /
• 2010

본 연구에서는 일련의 연구에서 수집된 영작문 오류 유형의 정제된 자료를 토대로 연관 규칙을 생성하고, 학습을 통해서 효용성이 검증된 연관 규칙을 활용해서 영작문 데이터의 형태 통사 오류를 자동으로 탐지한다. 영작문 데이터에서 형태 통사 오류를 찾아내는 작업은 많은 시간과 자원이 소요되는 작업이므로 자동화가 필수적이다. 기존의 연구들이 통계적 모델을 활용한 어휘적 오류에 치중하거나 언어 이론적 틀에 근거한 통사 처리에 집중하는 반면에, 본 연구는 데이터 마이닝을 통해서 정제된 데이터에서 연관 규칙을 생성하고 이를 검증한 후 형태 통사 오류를 감지한다. 이전 연구들에서는 이론적 틀에 맞추어진 규칙 생성이나 언어 모델 생성을 위한 대량의 코퍼스 데이터와 같은 다량의 지식 베이스 생성이 필수적인데, 본 연구는 적은 양의 정제된 데이터를 활용한다. 영작문 오류 유형의 형태 통사 연관 규칙을 생성하기 위해서 Apriori 알고리즘을 활용하였다. 알고리즘을 통해서 생성된 연관 규칙 중 잘못된 규칙이 생성될 가능성이 있으므로, 상관성 검정, 코사인 유사도와 같은 규칙 효용성의 통계적 검증을 활용해서 타당한 규칙만을 학습하였다. 이를 통해서 축적된 연관 규칙들을 영작문 오류를 자동으로 탐지하는 실험에 활용하였다.

• The Journal of Korean Society for Radiation Therapy
• /
• v.28 no.1
• /
• pp.27-34
• /
• 2016

Purpose : Retrospective evaluation of setup changes using the corrected position during helical tomotherapy Materials and Methods : Head and neck cancer patients were randomly sampled and summarized into 3 groups: Group 1(32) Brain, Group 2 2(28)Maxillar, Nasal cavity, Group 3 (35) Nasopharynx(NPX), Tongue, Tonsil, and Oropharynx(OPX). In 3 groups, the statistical tests based on repeated measurements among 30 times of the duration of treatment by applying X, Y, Z axis errors, roll, weight changes, and vectors as variables. Results : The statistical test results showed that there was no difference between x-axis (p = 0.458) and y-axis (p=0.986) and in roll (p = 0.037), weight change (p <0.001), and the vector (p <0.001). In addition, the pattern between the three groups based on the fraction revealed no difference in x-axis (p = 0.430) and roll (p = 0.299) but a difference in y-axis (.023), weight change (p = 0.001), and vector (p = 0.028). Conclusion : The results of the retrospective evaluation found the change in the group 3 with respect Y, Z, weight, and vector and a larger random error during the treatment including low neck.