• 대한핵의학회지
• /
• 제37권5호
• /
• pp.288-300
• /
• 2003

목적: ${H_2}^{15}O$ PET의 정량화를 위하여 1-조직 구획모델이 쓰이며, 뇌혈류와 조직/혈액 분배계수를 구하기 위하여 nonlinear least squares (NLS) 방법이 사용되나 계산 시간이 긴 등의 문제로 파라미터를 각화소마다 구해야 하는 파라메트릭 영상 구성에는 적합하지 않다. 이 연구에서는 이와 같은 NLS 문제점을 극복하여 파라메트릭 영상을 빠르게 구성하기 위하여 제안된 파라미터 추정 알고리즘들을 구현하고, 이 방법들의 통계적 신뢰도와 계산의 효율성을 비교하였다. 대상 및 방법: 이 연구에서 이용한 방법들은 linear least squares (LLS), linear weighted least squares (LWLS), linear generalized least squares (GLS), linear generalized weighted least squares (GWLS), weighted integration (WI), 그리고 model-based clustering method (CAKS)이다. 노이즈 정도에 따른 각 파라메트릭 영상법의 정확성 및 통계적 신뢰성을 알아보기 위하여 Zubal 뇌모형(brain phantom)으로부터 동적 PET 영상을 모사하고 포아송노이즈를 더한 후 각 파라메트릭 영상 구성 방법을 적용하였다. 또한 정상인 16명에 대하여 얻은 실제 자료에 대하여 이 방법들을 적용하고 결과를 비교하였다. 결과: 뇌혈류와 분배계수에 대한 평균 오차는 방법에 따라 크게 다르지 않았으며 모든 방법이 뇌혈류 및 분배계수 추정에 있어 무시할 만한 바이어스를 보였다. 파라메트릭 영상의 정성적 특성 또한 유사하였으나 CAKS 방법의 계산 속도가 월등하여 NLS 방법의 약 1/500, LLS 방법의 약 1/25의 계산시간을 보였다. 결론: 뇌혈류 파라메트릭 영상 구성을 위한 빠른 파라미터 추정 알고리즘들 중에 보다 개선되어 제안된 LWS, GLS, GLWS, CAKS 방법들이 단순하고 빠른 LLS, WI 방법들에 비하여 통계적 신뢰성을 크게 향상시키지는 못하나 CAKS 방법은 계산 시간을 유의하게 단축시키므로 가장 적합한 파라메트릭 영상 구성방법이라 할 수 있을 것이다.

• Journal of the Korean Data and Information Science Society
• /
• 제25권4호
• /
• pp.931-939
• /
• 2014

In this paper we propose the iteratively reweighted least squares procedure to solve the quadratic programming problem of support vector expectile regression with an asymmetrically weighted squares loss function. The proposed procedure enables us to select the appropriate hyperparameters easily by using the generalized cross validation function. Through numerical studies on the artificial and the real data sets we show the effectiveness of the proposed method on the estimation performances.

• 한국공작기계학회논문집
• /
• 제12권4호
• /
• pp.17-22
• /
• 2003

This paper presents a new circular curve fitting approach of articulated manipulators, based on least square. The approach aims at gaining the interpolation of circle from n data points, under the condition that the fitted circle should pass both a starting point and an ending point. First a spherical fitting should be performed, using least squares. Then the circular curve fitting can be resulted from the intersection of the fitted sphere and the plane obtained from 3 points, i. e., a starting point, an ending point and the center of a sphere. The proposed algorithms are shown to be efficient by using MATLAB-based simulation.

• 응용통계연구
• /
• 제28권6호
• /
• pp.1235-1243
• /
• 2015

최근 들어 ICT 분야의 발달에 따라 데이터 사용량의 급격한 증가로 인터넷 트래픽 사용량 예측은 중요성은 강조되고 있다. 이러한 예측치를 적절한 트래픽 관리와 제어를 위한 계획 수립에 도움을 준다. 본 논문은, 5분 단위의 인터넷 트래픽 자료를 이용하여 결합 예측 모형을 제안하고자 한다. 이에 대하여 시계열의 대표적인 3개 모형인 Seasonal ARIMA, Fractional ARIMA(FARIMA), Taylor의 수정된 Holt-Winters 모형을 적용하였다. 모형 간 결합 예측 방법으로 예측치 간의 SA(Simple Average) 결합 예측 방법과 OLS(Ordinary Least Square)를 이용한 결합방법, ERLS(Equality Restricted Least Squares)를 이용한 결합 예측 방법, Armstrong(2001)이 제안한 MSE 기반 결합 예측 방법을 사용한다. 이에 따른 결과로서 3시간에서의 예측은 Seasonal ARIMA가 선택된 반면, 6시간 이후 예측에서는 결합 예측 방법이 좋은 예측 성능을 보여준다.

• Bulletin of the Korean Chemical Society
• /
• 제25권5호
• /
• pp.647-651
• /
• 2004

Partial Least Squares (PLS) calibration performance has been systematically investigated by changing spectral resolutions of near-infrared (NIR) spectra. For this purpose, synthetic samples simulating naphtha were prepared to examine the calibration performance in complex chemical matrix. These samples were composed of $C_6-C_9$ normal paraffin, iso-paraffin, naphthene, and aromatic hydrocarbons. NIR spectra with four different resolutions of 4, 8, 16, and 32$cm^{-1}$ were collected and then PLS regression was performed. For PLS calibration, five different group compositions (such as total paraffin content) and six different pure components (such as benzene concentration) were selected. The overall results showed that at least 8$cm^{-1}$ resolution was required to resolve the complex chemical matrix such as naphtha. It was found that the influence of resolution on the PLS calibration was varied by the spectral features of a component.

• Archives of Pharmacal Research
• /
• 제28권5호
• /
• pp.598-603
• /
• 2005

The purpose of this study was to determine the influence of weight with gentamicin assay error on the Bayesian and nonlinear least squares regression analysis in 12 Korean appen dicitis patients. Gentamicin was administered intravenously over 0.5 h every 8 h. Three specimens were collected at 48 h after the first dose from all patients at the following times, just before regularly scheduled infusion, at 0.5 h and 2 h after the end of 0.5 h infusion. Serum gentamicin levels were analyzed by fluorescence polarization immunoassay technique with TDxFLx. The standard deviation (SD) of the assay over its working range had been determined at the serum gentamicin concentrations of 0, 2, 4, 8, 12, and 16 ${\mu}g$/mL in quadruplicate. The polynominal equation of gentamicin assay error was found to be SD (${\mu}g$/mL) = 0.0246-(0.0495C)+ (0.00203C$^2$). There were differences in the influence of weight with gentamicin assay error on pharmacokinetic parameters of gentamicin using the nonlinear least squares regression analysis but there were no differences on the Bayesian analysis. This polynominal equation can be used to improve the precision of fitting of pharmacokinetic models to optimize the process of model simulation both for population and for individualized pharmacokinetic models. The result would be improved dosage regimens and better, safer care of patients receiving gentamicin.

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 2018년도 전력전자학술대회
• /
• pp.237-239
• /
• 2018

This paper proposes an on-line estimation algorithm of internal resistance of Li-ion battery based on the recursive least squares method to prevent the overvoltage and undervoltage casing degradation of life cycle of battery. An equivalent circuit model with single time constant is adopted, and under assumptions that the terminal voltage, current and SOC are measured accurately, the discrete time based nonlinear equation of the model can be converted to the linear equation which can be applied to recursive least squares method. Since the coefficients of the discrete time linear equation can be expressed by the parameters of the equivalent circuit model, it is shown that an internal resistance (Ri) can be estimated in real time using the least square method.

• Communications for Statistical Applications and Methods
• /
• 제15권2호
• /
• pp.205-211
• /
• 2008

This paper deals with the estimations of kernel ridge regression when the responses are subject to randomly right censoring. The iterative reweighted least squares(IRWLS) procedure is employed to treat censored observations. The hyperparameters of model which affect the performance of the proposed procedure are selected by a generalized cross validation(GCV) function. Experimental results are then presented which indicate the performance of the proposed procedure.

• Journal of the Korean Data and Information Science Society
• /
• 제19권2호
• /
• pp.577-585
• /
• 2008

This paper suggests a marginal probability model for analyzing repeated polytomous response data when some factors are nested in others in treatment structures on a larger experimental unit. As a repeated measures factor, time is considered on a smaller experimental unit. So, two different experiment sizes are considered. Each size of experimental unit has its own design structure and treatment structure, and the marginal probability model can be constructed from the structures for each size of experimental unit. Weighted least squares(WLS) methods are used for estimating fixed effects in the suggested model.

• Communications for Statistical Applications and Methods
• /
• 제19권3호
• /
• pp.423-432
• /
• 2012

This paper is concerned with the detection of multiple change-points in linear regression models. The proposed procedure relies on the local estimation for global change-point estimation. We propose a multiple change-point estimator based on the local least squares estimators for the regression coefficients and the split measure when the number of change-points is unknown. Its statistical properties are shown and its performance is assessed by simulations and real data applications.