• Journal of the Korean Data and Information Science Society
• /
• 제17권2호
• /
• pp.401-409
• /
• 2006

We consider a generalized partly-parametric additive risk model which generalizes the partly parametric additive risk model suggested by McKeague and Sasieni (1994). As an estimation method of this model, we propose to use the weighted least square estimation, suggested by Huffer and McKeague (1991), for Aalen's additive risk model by a piecewise constant risk. We provide an illustrative example as well as a simulation study that compares the performance of our method with the ordinary least squares method.

• 한국전산구조공학회논문집
• /
• 제20권6호
• /
• pp.779-787
• /
• 2007

본 연구는 계면경계에서 특이성을 갖는 이종재료 열전달문제를 효율적으로 해석할 수 있는 이동최소제곱 유한차분법을 제시한다 이동최소제곱 유한차분법은 격자망(grid)없이 절점만으로 이동최소제곱법을 이용하여 Taylor 다항식을 구성하고 차분식을 만들어 미분방정식을 직접 푼다. 초평면함수 개념에 근거한 쐐기함수를 이동최소제곱 센스(sense)로 근사식에 매입하여 쐐기거동과 미분 점프에 따른 계면경계 특성을 효과적으로 묘사하고 고속으로 미분을 근사하는 이동최소제곱 유한차분법의 강점을 발휘하도록 했다. 서로 다른 열전달계수를 갖는 이종재료 열전도문제 해석을 통해 이동최소제곱 유한차분법이 계면경계문제에서도 뛰어난 계산효율성과 해의 정확성을 확보할 수 있음을 보였다.

• Journal of the Korean Data and Information Science Society
• /
• 제22권5호
• /
• pp.931-940
• /
• 2011

분류분석은 학습표본으로부터 분류규칙을 도출한 후 새로운 표본에 적용하여 특정 범주로 분류하는 방법이다. 데이터의 복잡성에 따라 다양한 분류분석 방법이 개발되어 왔지만, 데이터 차원이 높고 변수간 상관성이 높은 경우 정확하게 분류하는 것은 쉽지 않다. 본 연구에서는 데이터차원이 상대적으로 높고 변수간 상관성이 높을 때 강건한 분류방법을 제안하고자 한다. 부분최소자승법은 연속형데이터에 사용되는 기법으로서 고차원이면서 독립변수간 상관성이 높을 때 예측력이 높은 통계기법으로 알려져 있는 다변량 분석기법이다. 벌점 부분최소자승법을 이용한 분류방법을 실제데이터와 시뮬레이션을 적용하여 성능을 비교하고자 한다.

• 대한토목학회논문집
• /
• 제8권2호
• /
• pp.177-184
• /
• 1988

지금까지 관측(觀測)값의 처리(處理)에 적용(適用)해 왔던 최소(最小)제곱법(法)은 정적(靜的)인 data의 처리(處理)에는 적합(適合)하나, 시간(時間)의 변화(變化)가 수반(隨伴)된 관측치(觀測値)의 처리(處理)에는 부적당(不適當)하다. 본(本) 연구(硏究)는 모든 관측(觀測)값이 시간(時間)의 함수(函數)인 것을 고려(考慮)해서 종래(從來)의 최소(最小)제곱법(法)의 일반식(一般式)에 시간(時間)의 개념(槪念)을 도입(導入)하여 동적(動的)으로 수준망(水準網)을 해석(解析)하였다. 이 경우(境遇) 관측(觀測) data를 동적최소(動的最小)제곱법(法)으로 처리(處理)한 결과(結果) 정밀(精密)한 조정(調整)값을 얻었다.

• Structural Engineering and Mechanics
• /
• 제22권3호
• /
• pp.331-349
• /
• 2006

The stresses and deflections in a laminated rectangular plate under thermal vibration are determined by using the moving least squares differential quadrature (MLSDQ) method based on the first order shear deformation theory. The weighting coefficients used in MLSDQ approximation are obtained through a fast computation of the MLS shape functions and their partial derivatives. By using this method, the governing differential equations are transformed into sets of linear homogeneous algebraic equations in terms of the displacement components at each discrete point. Boundary conditions are implemented through discrete grid points by constraining displacements, bending moments and rotations of the plate. Solving this set of algebraic equations yields the displacement components. Then substituting these displacements into the constitutive equation, we obtain the stresses. The approximate solutions for stress and deflection of laminated plate with cross layer under thermal load are obtained. Numerical results show that the MLSDQ method provides rapidly convergent and accurate solutions for calculating the stresses and deflections in a multi-layered plate of cross ply laminate subjected to thermal vibration of sinusoidal temperature including shear deformation with a few grid points.

• Communications for Statistical Applications and Methods
• /
• 제10권3호
• /
• pp.873-878
• /
• 2003

LS-SVM(Least-Squares Support Vector Machine) has been used as a promising method for regression as well as classification. Suykens et al.(2000) used only the magnitude of residuals to obtain SVs(Support Vectors). Suykens' method behaves well for homogeneous model. But in a heteroscedastic model, the method shows a poor behavior. The present paper proposes a new method to get SVs. The proposed method uses the variance of noise as well as the magnitude of residuals to obtain support vectors. Through the simulation study we justified excellence of our proposed method.

• Journal of Advanced Marine Engineering and Technology
• /
• 제20권2호
• /
• pp.108-108
• /
• 1996

This paper describes a new approach to solve balancing of a rigid rotor. In this paper, the balancing of the rigid rotor using genetic algorithms, which are search algorithms based on the mechanics of natural selection and natural genetics is proposed. Under the assumption that the initial vibration values used to calculate correction masses contain errors, the influence coefficient method, the least squares method and a genetic algorithm are compared. The results show that the vibration amplitude obtained with the least squares method and the genetic algorithm is smaller than that obtained with the influence coefficient method.

• Journal of Advanced Marine Engineering and Technology
• /
• 제20권2호
• /
• pp.40-47
• /
• 1996

This paper describes a new approach to solve balancing of a rigid rotor. In this paper, the balancing of the rigid rotor using genetic algorithms, which are search algorithms based on the mechanics of natural selection and natural genetics is proposed. Under the assumption that the initial vibration values used to calculate correction masses contain errors, the influence coefficient method, the least squares method and a genetic algorithm are compared. The results show that the vibration amplitude obtained with the least squares method and the genetic algorithm is smaller than that obtained with the influence coefficient method.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1694-1700
• /
• 2004

In this paper, we propose the AWLS/MFT (Adaptive Weighed Least Squares/ Modulation Function Technique) devised by A. E. Pearson et al. for the transfer function estimation of a modal system and investigate the performance of several algorithms, the Gram matrix method, a Luenberger Observer (LO), Least Squares (LS), and Recursive Least Squares (RLS), for the estimation of initial conditions. With the benefit of the Modulation Function Technique (MFT), we can separate the estimation problem into two phases: the transfer function parameters are estimated in the first phase, and the initial conditions are estimated in the second phase. The LO method produces excellent IC estimates in the noise free case, but the other three methods show better performance in the noisy case. Finally, we compared our result with the Prony based method. In the noisy case, the AWLS and one of the three methods - Gram matrix, LS, and RLS- show better performance in the output Signal to Error Ratio (SER) aspect than the Prony based method under the same simulation conditions.

• Communications for Statistical Applications and Methods
• /
• 제15권3호
• /
• pp.441-450
• /
• 2008

In this paper we propose a new method for solving multiclass problem with least squares support vector machine(LS-SVM) regression. This method implements one-against-all scheme which is as accurate as any other approach. We also propose cross validation(CV) method to select effectively the optimal values of hyper-parameters which affect the performance of the proposed multiclass method. Experimental results are then presented which indicate the performance of the proposed multiclass method.