• 응용통계연구
• /
• 제21권2호
• /
• pp.275-290
• /
• 2008

마이크로어레이 유전자 발현 자료는 대용량이며 또한 관측 과정이 복잡하여 결측치가 빈번하게 발생된다. 본 논문에서는 관측 시점 간에 상관성을 갖는 시간경로 유전자 발현 자료에 대한 결측치 추정을 위하여 순차적 부분최소제곱(sequential partial least squares: SPLS) 회귀적합 방법을 제안한다. 이는 순차적 기법과 부분최소제곱(partial least squares: PLS) 회귀적합 방법을 결합시킨 것이다. 세 가지의 이스트(yeast) 시간경로 자료들에 대한 몇 가지 모의실험을 통하여 제안된 결측치 추정방법의 유용성을 평가한다.

• 한국음향학회지
• /
• 제40권3호
• /
• pp.213-218
• /
• 2021

입력과 출력에 동시에 잡음이 존재하는 경우 최소 자승법 보다는 완전 최소 자승법이 더 우수한 추정 성능을 보인다는 것이 알려져 있다. 완전 최소 자승법을 시계열 특성을 가지는 데이터에 적용할 경우 보다 실시간 성을 더하기 위해서 Recursive Total Least Squares(RTS) 알고리즘이 제안되어 있다. RTLS는 알고리즘 내에 존재하는 역행렬 계산에서 수치적인 불안정성을 지닌다. 본 논문에서는 RTLS와 유사한 수렴성을 지닐 뿐만 아니라 수치적 불안정성을 줄이기 위한 알고리즘을 제안한다. 이 알고리즘을 위해서 Iterative Wiener Filter(IWF)를 적용한 새로운 RTLS를 제안한다. 시뮬레이션을 통해서 수렴성이 기존의 RTLS와 유사할 뿐만 아니라 수치적 견실성이 기존 RTLS보다 향상되었다는 것을 보인다.

• East Asian mathematical journal
• /
• 제15권1호
• /
• pp.121-133
• /
• 1999

• Communications for Statistical Applications and Methods
• /
• 제25권6호
• /
• pp.673-685
• /
• 2018

Procrustes analysis is a useful technique useful to measure, compare shape differences and estimate a mean shape for objects; however it is based on a least squares criterion and is affected by some outliers. Therefore, we propose two generalized Procrustes analysis methods based on M-estimation and least median of squares estimation that are resistant to object outliers. In addition, two algorithms are given for practical implementation. A simulation study and some examples are used to examine and compared the performances of the algorithms with the least square method. Moreover since these resistant GPA methods are available for higher dimensions, we need some methods to visualize the objects and mean shape effectively. Also since we have concentrated on resistant fitting methods without considering shape distributions, we wish to shape analysis not be sensitive to particular model.

• Journal of Electrical Engineering and Technology
• /
• 제7권4호
• /
• pp.606-614
• /
• 2012

A practical recursive linear robust estimation scheme is proposed for target localization in the sensor network which provides range difference of arrival (RDOA) measurements. In order to radically solve the known practical difficulties such as sensitivity for initial guess and heavy computational burden caused by intrinsic nonlinearity of the RDOA based target localization problem, an uncertain linear measurement model is newly derived. In the suggested problem setting, the target localization performance of the conventional linear estimation schemes might be severely degraded under the low SNR condition and be affected by the target position in the sensor network. This motivates us to devise a new sensor network localization algorithm within the framework of the recently developed robust least squares estimation theory. Provided that the statistical information regarding RDOA measurements are available, the estimate of the proposition method shows the convergence in probability to the true target position. Through the computer simulations, the omnidirectional target localization performance and consistency of the proposed algorithm are compared to those of the existing ones. It is shown that the proposed method is more reliable than the total least squares method and the linear correction least squares method.

• International Journal of Aeronautical and Space Sciences
• /
• 제14권1호
• /
• pp.85-90
• /
• 2013

This paper presents a study on terrain referenced navigation (TRN). The extended Kalman filter (EKF) is adopted as a filter method. A Jacobian matrix of measurement equations in the EKF consists of terrain slope terms, and accurate slope estimation is essential to keep filter stability. Two slope estimation methods are proposed in this study. Both methods are based on the least-squares approach. One is planar regression searching the best plane, in the least-squares sense, representing the terrain map over the region, determined by position error covariance. It is shown that the method could provide a more accurate solution than the previously developed linear regression approach, which uses lines rather than a plane in the least-squares measure. The other proposed method is weighted planar regression. Additional weights formed by Gaussian pdf are multiplied in the planar regression, to reflect the actual pdf of the position estimate of EKF. Monte Carlo simulations are conducted, to compare the performance between the previous and two proposed methods, by analyzing the filter properties of divergence probability and convergence speed. It is expected that one of the slope estimation methods could be implemented, after determining which of the filter properties is more significant at each mission.

• 조명전기설비학회논문지
• /
• 제21권7호
• /
• pp.61-67
• /
• 2007

전력수요예측은 전력계통의 운용을 위해 필수적이다. 따라서 다양한 방법이 제시되어 왔으며, 특히 특수일의 수요예측은 평일과 구분되며, 부하 패턴을 축출하기에 충분한 자료 확보가 어려워 예측 오차가 크게 나타난다. 본 논문에서는 특수일의 부하예측 정확도를 개선하기 위해 퍼지 최소자승 선형회귀 모델을 분석한다. 4종류의 퍼지 최소자승 선형회귀 모델에 대해 분석과 사례연구를 통하여 가장 정확한 모델을 제시한다.

• 한국해양공학회지
• /
• 제24권1호
• /
• pp.20-33
• /
• 2010

The paper deals with strength design of a riser support installed on floating production storage and offloading (FPSO) vessel under various loading conditions - operation, extreme, damaged, one line failure case (OLFC) and installation. The design problem is formulated such that thickness sizing variables are determined by minimizing the weight of a riser support structure subject to stresses constraints. The initial design model is generated based on an actual FPSO riser support specification. The finite element analysis (FEA) is conducted using MSC/NASTRAN, and optimal solutions are obtained via moving least squares method (MLSM) in the context of response surface based approximate optimization. For the meta-modeling of inequality constraint functions of stresses, a constraint-feasible moving least squares method (CF-MLSM) is used in the present study. The method of CF-MLSM, compared to a conventional MLSM, has been shown to ensure the constraint feasibility in a case where the approximate optimization process is employed. The optimization results present improved design performances under various riser operating conditions.

• 한국음향학회지
• /
• 제15권4호
• /
• pp.93-97
• /
• 1996

완전최소자승법(total least squares method, TLS) Ax${\simeq}$b와 같은 형태의 시스템 식을 푸는데 있어 데이터 행렬 A와 b에 잡음비 섞인 경우에 편이 되지 않은 해를 구하기 위하여 널리 이용된다. 그러나 임펄수성의 잡음과 같은 heavy tailed 확률분포를 갖는 잡음이 존재할 때 완전 최소자승법은 unbiased estimator이지만 최소자승법(least squares, LS)과 마찬가지로 경인하지 못한 성능을 보인다. 본 논문에서는 TLS 방법의 견인성에 대하여 논하고 완전최소자승법의 해의 특성을 기반으로 하여 견인한 완전최소자승법(robust TLS, ROTLS)을 제안한다. 또한 ROTLS 방법을 시스템식별문제에 적용하여 그 성능을 평가한다.

• Near Infrared Analysis
• /
• 제1권2호
• /
• pp.1-8
• /
• 2000

Using the net analyte signal, hybrid linear analysis was proposed to predict chemical concentration. In this paper, we select a sample from training set and apply orthogonal signal correction to obtain an improved pseudo unit spectrum for hybrid least analysis. using the mean spectrum of a calibration training set, we first show the calibration by hybrid least analysis is effective to the prediction of not only chemical concentrations but also physical property variables. Then, a pseudo unit spectrum from a training set is also tested with and without orthogonal signal correction. We use two data sets, one including five chemical concentrations and the other including ten physical property variables, to compare the performance of partial least squares and modified hybrid least analysis calibration methods. The results show that the hybrid least analysis with a selected training spectrum instead of well-measured pure spectrum still gives good performances, which is a little better than partial least squares.