• Korean Journal of Hydrosciences
• /
• 제3권
• /
• pp.105-113
• /
• 1992

Various modeling approaches to study a long term behavior of streamflow or groundwater storage have been conducted. In this study, a Multivariate AR (1) Model has been applied to generate monthly flows of the one key station which has historical flows using monthly flows of the three subordinate stations. The Model performance was examined using statistical comparisons between the historical and generated monthly series such as mean, variance, skewness. Also, the correlation coefficients (lag-zero, and lag-one) between the two monthly flows were compared. The results showed that the modeled generated flows were statistically similar to the historical flows.

• Communications for Statistical Applications and Methods
• /
• 제27권2호
• /
• pp.255-268
• /
• 2020

A mixture of multivariate canonical fundamental skew t-distribution (CFUST) has been of interest in various fields. In particular, interest in the unsupervised learning society is noteworthy. However, fitting the model via EM algorithm suffers from significant processing time. The main cause is due to the calculation of many multivariate t-cdfs (cumulative distribution functions) in E-step. In this article, we provide an approximate, but fast calculation method for the in univariate fashion, which is the product of successively conditional univariate t-cdfs with Taylor's first order approximation. By replacing all multivariate t-cdfs in E-step with the proposed approximate versions, we obtain the admissible results of fitting the model, where it gives 85% reduction time for the 5 dimensional skewness case of the Australian Institution Sport data set. For this approach, discussions about rough properties, advantages and limits are also presented.

• Journal of the Korean Statistical Society
• /
• 제34권2호
• /
• pp.109-123
• /
• 2005

Variogram estimation is an important step of spatial statistics since it determines the kriging weights. Matheron's variogram estimator can be written as a quadratic form of the observed data. In this paper, we extend a skew t distribution to a generalized skew t distribution and moments of the variogram estimator for a generalized skew t distribution are derived in closed forms. After calculating the correlation structure of the variogram estimator, variogram fitting by generalized least squares is discussed.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.1108-1113
• /
• 2000

A condition monitoring methodology for rotating machinery is proposed based on multivariate statistical analysis. The CMS usually are using the vibration signal amplitude such as acceleration RMS, peak and velocity RMS to detect machine faults but the information is not so enough that CMS cannot perform reliable monitoring. So new parameters are added such as shape factor, crest factor, kurtosis and skewness as time domain parameters and spectrum amplitude of rotating frequency, $2^{nd}$ harmonics and gear mesh frequency etc. as frequency domain parameters. Many parameters are combined to represent the machine state using the Hotelling's $T^2$ statistics. The proposed methodology is tested in laboratory and the on-line experiment has shown that the proposed methodology offers a reliable monitoring for rotating machinery.

• 한국수자원학회논문집
• /
• 제30권2호
• /
• pp.155-163
• /
• 1997

본 연구에서는 낙동강유역 진동, 현풍, 왜관 지점의 연평균 유량자료에 대하여 다변량 추계학적 모형올 적용하여 가뭄특성을 해석하였다. 추계학적 모형으로는 다변량 자기회귀 (MAR) 모형과 다변량 contemporaneous 자기회귀 (MCAR) 모형올 사용하였으며, 잔차계열의 왜곡도 검사, 계열상관도(correlogram) 등의 적합도 검정을 통하여 MCAR(1) 모형과 MAR(1) 모형올 적정 모형으로 선정하였다. 또한 MCAR(1) 모형과 MAR(1) 모형에 의해 모의발생된 자료 모두 실제자료의 기본적인 통계값과 매우 비슷하게 나타났다. 따라서 모의발생된 다양한 크기의 자료를 통하여 산정된 3개 지점의 재현기간별 가뭄특성치, 예를 들변 가뭄기간, 가뭄부족량, 가뭄강도 둥은 비교적 잘 재현된 것으로 판단된다. 위와 같이 산정된 가뭄특성치는 중.장기간 수자원 공급체계를 위한 계획과 설계에 중요한 정보를 제공할 것으로 기대된다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2008년도 학술발표회 논문집
• /
• pp.694-698
• /
• 2008

In hydrology, it is a difficult task to deal with multivariate time series such as modeling streamflows of an entire complex river system. Normal distribution based model such as MARMA (Multivariate Autorgressive Moving average) has been a major approach for modeling the multivariate time series. There are some limitations for the normal based models. One of them might be the unfavorable data-transformation forcing that the data follow the normal distribution. Furthermore, the high dimension multivariate model requires the very large parameter matrix. As an alternative, one might be decomposing the multivariate data into independent components and modeling it individually. In 1985, Lins used Principal Component Analysis (PCA). The five scores, the decomposed data from the original data, were taken and were formulated individually. The one of the five scores were modeled with AR-2 while the others are modeled with AR-1 model. From the time series analysis using the scores of the five components, he noted "principal component time series might provide a relatively simple and meaningful alternative to conventional large MARMA models". This study is inspired from the researcher's quote to develop a multivariate simulation model. The multivariate simulation model is suggested here using Principal Component Analysis (PCA) and Independent Component Analysis (ICA). Three modeling step is applied for simulation. (1) PCA is used to decompose the correlated multivariate data into the uncorrelated data while ICA decomposes the data into independent components. Here, the autocorrelation structure of the decomposed data is still dominant, which is inherited from the data of the original domain. (2) Each component is resampled by block bootstrapping or K-nearest neighbor. (3) The resampled components bring back to original domain. From using the suggested approach one might expect that a) the simulated data are different with the historical data, b) no data transformation is required (in case of ICA), c) a complex system can be decomposed into independent component and modeled individually. The model with PCA and ICA are compared with the various statistics such as the basic statistics (mean, standard deviation, skewness, autocorrelation), and reservoir-related statistics, kernel density estimate.

• 한국방재학회 논문집
• /
• 제1권2호
• /
• pp.115-121
• /
• 2001

실측자료가 충분하지 못한 단기간의 유출량 자료로부터 추계학적 모형에 의해 장기간의 자료를 모의발생시키는 목적은 수공구조물의 설계에 필요한 설계홍수량의 산정 및 수자원 시스템의 운영조작 방침을 결정하기 위한 풍부한 입력자료를 제공하는데 있다. 특히 본 연구에서는 단일지점이 아닌 다지점에 대한 지점간 서로의 연관성을 고려한 하천유량의 추계학적인 모의 발생기법인 다변량 자기회귀 모형을 적용하고자 한다. 따라서 본 연구에서는 낙동강유역의 2개 지점에 대하여 다변량 모형을 적용하여 모의 발생된 월유량과 실측치를 통계적으로 비교, 분석하였다. 모의발생된 월유량과 실측치를 평균, 분산, 왜곡도, 상관관계 등에 의해 비교, 분석한 결과 모의발생된 월유량과 실측치는 통계적으로 매우 유사하게 나타났다.

• 물과 미래
• /
• 제24권4호
• /
• pp.67-72
• /
• 1991

단일지점(Single Site)에 대한 하천유량의 추계학적인 모의 발생을 위해서는 간단한 모델중의 하나로 Univariate AR(1) 모델이 흔히 쓰여왔다. 그러나 다지점(Multi Sites)에 대한 하천유량에 관한 추계학적인 모의발생은 지점간 서로의 연관성 때문에 단일지점을 위한 모의발생처럼 쉽게 해결되지 않았다. 본 연구에서는 미국 아이다호주의 Camas Creek 유역에 대하여 하나의 키이지점(Key Station)과 주변에 세 개의 종속지점(Subordinate Station)을 설정하고 다변량 AR(1)모델을 적용하여 모의발생된 월유량과 실측치를 통계적으로 비교, 분석하였다. 모의 발생된 월유량과 실측치를 평균, 분산, 왜곡도, 상관관계등에 의해 비교, 분석한 결과 모이 발생된 월유량과 실측치는 통계적으로 서로 유사성을 보였다.

• Investigative Magnetic Resonance Imaging
• /
• 제25권2호
• /
• pp.118-128
• /
• 2021

Purpose: To investigate the value of MR textural analysis, including use of diffusion-weighted imaging (DWI) to differentiate malignant from benign soft-tissue tumors on 3T MRI. Materials and Methods: We enrolled 69 patients (25 men, 44 women, ages 18 to 84 years) with pathologically confirmed soft-tissue tumors (29 benign, 40 malignant) who underwent pre-treatment 3T-MRI. We calculated MR texture, including mean, standard deviation (SD), skewness, kurtosis, mean of positive pixels (MPP), and entropy, according to different spatial-scale factors (SSF, 0, 2, 4, 6) on axial T1- and T2-weighted images (T1WI, T2WI), contrast-enhanced T1WI (CE-T1WI), high b-value DWI (800 sec/mm²), and apparent diffusion coefficient (ADC) map. We used the Mann-Whitney U test, logistic regression, and area under the receiver operating characteristic curve (AUC) for statistical analysis. Results: Malignant soft-tissue tumors had significantly lower mean values of DWI, ADC, T2WI and CE-T1WI, MPP of ADC, and CE-T1WI, but significantly higher kurtosis of DWI, T1WI, and CE-T1WI, and entropy of DWI, ADC, and T2WI than did benign tumors (P < 0.050). In multivariate logistic regression, the mean ADC value (SSF, 6) and kurtosis of CE-T1WI (SSF, 4) were independently associated with malignancy (P ≤ 0.009). A multivariate model of MR features worked well for diagnosis of malignant soft-tissue tumors (AUC, 0.909). Conclusion: Accurate diagnosis could be obtained using MR textural analysis with DWI and CE-T1WI in differentiating benign from malignant soft-tissue tumors.

• Journal of Periodontal and Implant Science
• /
• 제36권3호
• /
• pp.591-600
• /
• 2006

Purpose: This study assessed the impact of self-reported periodontal health on the oral health-related quality of life among elderly Koreans. Methods: Four hundred twenty one elderly Koreans in Seoul and suburban areas were selected with a cluster (institution) sampling method, and were requested to take oral examinations and finish questionnaires on the Oral Health Impact Profile-14(OHIP-14). and self-reported periodontal health status, such as periodontal symptoms, self-rated periodontal health and periodontal treatment need. As the dependent variable, OHIP-14 showed a positive skewed distribution (skewness: 1.17), we transformed to square-root form to apply parametric analyses. Bivariate analysis by t-test and ANOVA, and multivariate analysis with the two-level regression model accounting clusters were implemented. Results: Mean age of the subjects was 74.6 years and 66.5% were women. Fourteen items of OHIP-14 were summarized to one factor explaining 78.6% of total variance and produced the Chronbach alpha coefficient of 0.92. Results from the multivariate model, adjusting for age, sex, type of institutions, ability to pay, and number of teeth present, showed significantly lower OHIP-14 with reporting less than 3 periodontal symptoms (p(O.OOO1), rating their own periodontal health as above average level (p=O.0144), and thinking they don't need any periodontal treatments in the near future (p=O.0148), than their counterparts. The intraclass-corrrelation estimated by the final model was 0.028. Conclusion: This study demonstrates a significant association between self-reported periodontal health status and the oral health-related quality of life.