• Communications for Statistical Applications and Methods
• /
• v.27 no.2
• /
• 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.

• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.187-191
• /
• 2000

In the analysis of repeated measurements, multivariate regression models that account for the correlations among the observations from the same subject are widely used. Like the usual univariate regression models, these multivariate regression models also need some model diagnostic procedures. In this paper, we propose a simple graphical method to detect outliers and to investigate the goodness of model fit in repeated measures data. The graphical method is based on the quantile-quantile(Q-Q) plots of the $X^2$ distribution and the standard normal distribution. We also propose diagnostic measures to detect influential observations. The proposed method is illustrated using two examples.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.3
• /
• pp.421-427
• /
• 2013

The dependence between wind speeds in multiple wind sites has a considerable impact on the reliability of power systems containing wind energy. This paper presents a new method to generate dependent wind speed time series (WSTS) based on copulas theory. The basic feature of the method lies in separating multivariate WSTS into dependence structure and univariate time series. The dependence structure is modeled through the use of copulas, which, unlike the cross-correlation matrix, give a complete description of the joint distribution. An autoregressive moving average (ARMA) model is applied to represent univariate time series of wind speed. The proposed model is illustrated using wind data from two sites in Canada. The IEEE Reliability Test System (IEEE-RTS) is used to examine the proposed model and the impact of wind speed dependence between different wind regimes on the generation system reliability. The results confirm that the wind speed dependence has a negative effect on the generation system reliability.

• Communications for Statistical Applications and Methods
• /
• v.11 no.1
• /
• pp.153-160
• /
• 2004

In recent years, the size of data set which we usually handle is enormous, so a lot of outliers could be included in data set. Therefore the robust procedures that automatically handle outliers become very importance issue. We consider the robust estimation problem of location parameter in the univariate case. In this paper, we propose a new method for defining robustness weights for the weighted mean based on the median distance of observations and compare its performance with several existing robust estimators by a simulation study. It turns out that the proposed method is very competitive.

• Journal of Fashion Business
• /
• v.15 no.6
• /
• pp.176-203
• /
• 2011

In today's intensifying global competition, Korean fashion industry is relying on only qualitative data for feasibility study of future projects and developmental plan. This study was conducted in order to support establishment of a scientific and rational management system that reflects market demand. First, fashion market size was limited to the total amount of expenditure for fashion clothing products directly purchased by Koreans for wear during 6 months in spring and summer and 6 months in autumn and winter. Fashion market forecasting model was developed using statistical forecasting method proposed by previous research. Specifically, time series model was selected, which is a verified statistical forecasting method that can predict future demand when data from the past is available. The time series for empirical analysis was fashion market sizes for 8 segmented markets at 22 time points, obtained twice each year by the author from 1998 to 2008. Targets of the demand forecasting model were 21 research models: total of 7 markets (excluding outerwear market which is sensitive to seasonal index), including 6 segmented markets (men's formal wear, women's formal wear, casual wear, sportswear, underwear, and children's wear) and the total market, and these markets were divided in time into the first half, the second half, and the whole year. To develop demand forecasting model, time series of the 21 research targets were used to develop univariate time series models using 9 types of exponential smoothing methods. The forecasting models predicted the demands in most fashion markets to grow, but demand for women's formal wear market was forecasted to decrease. Decrease in demand for women's formal wear market has been pronounced since 2002 when casualization of fashion market intensified, and this trend was analyzed to continue affecting the demand in the future.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.3 no.1
• /
• pp.81-98
• /
• 1999

In spite of the well developed theory and the practical use of the univariate B-spline, the theory of multivariate B-spline is very new and waits its practical use. We compare in this article the multivariate B-spline approximation with the polynomial approximation for the surface fitting. The graphical and numerical comparisons show that the multivariate B-spline approximation gives much better fitting than the polynomial one, especially for the surfaces which vary very rapidly.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.10 no.2
• /
• pp.38-44
• /
• 1985

Systems reliability can be improved by using redundancy and/or developing more reliable components. This papaer considers a joint optimization of both alternatives for a series system. It is shown that the n-stage optimization problem can be decomposed into n single stage subproblem. Each subproblem is further transformed into a univariate optimization problem for which a simple and efficient solution method is developed.

• Journal of the Korean Statistical Society
• /
• v.33 no.4
• /
• pp.435-447
• /
• 2004

In additive nonparametric regression, Linton and Nielsen (1995) showed that the marginal integration when applied to the local linear smoother produces a rate-optimal estimator of each univariate component function for the case where the dimension of the predictor is two. In this paper we give new formulas for the bias and variance of the marginal integration regression estimators which are valid for boundary areas as well as fixed interior points, and show the local linear marginal integration estimator is in fact rate-optimal when the dimension of the predictor is less than or equal to four. We extend the results to the case of the local polynomial smoother, too.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2009.10a
• /
• pp.113-118
• /
• 2009

A designer regards the vibration system as a linear system. However, in real world, nonlinearity of a vibration system should exist caused by various factors like manufacturing conditions or uncertain material properties. So, properties of a spring and a damper which are consisting the vibration system have statistical distribution. Therefore, a designer needs to analyze the statistical nonlinearity in a vibration system. In this paper, $1^{st}$ Taylor series expansion method and univariate dimension reduction method apply to a performance measure of nonlinear vibration system, and compare each result. And then, merits and demerits of each method are discussed. For apply more actual problem, a performance measure population is estimated based on design variable samples like properties of spring or damper.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2612-2616
• /
• 2003

Method for performance monitoring and diagnosis of a MIMO control system has been studied aiming at application to model predictive control (MPC) for industrial processes. The performance monitoring part is designed on the basis of the traditional SPC/SQC method. To meet the underlying premise of Schwart chart observation that the observed variable should be univariate and independent, the process variables are decorrelated temporally as well as spatially before monitoring. The diagnosis part was designed to identify the root of performance degradation among the controller, process, and disturbance. For this, a method to estimate the model-error and disturbance signal has been devised. The proposed methods were evaluated through numerical examples.