http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Regression with Censored Data by Least Squares Support Vector Machine
Daehak Kim,Jooyong Shin,Kwangsik Oh 한국통계학회 2004 Journal of the Korean Statistical Society Vol.33 No.1
In this paper we propose a prediction method on the regression modelwith randomly censored observations of the training data set. The leastsquares support vector machine regression is applied for the regression func-tion prediction by incorporating the weights assessed upon each observationin the optimization problem. Numerical examples are given to show theperformance of the proposed prediction method.
On Line LS-SVM for Classification
Kim, Daehak,Oh, KwangSik,Shim, Jooyong The Korean Statistical Society 2003 Communications for statistical applications and me Vol.10 No.2
In this paper we propose an on line training method for classification based on least squares support vector machine. Proposed method enables the computation cost to be reduced and the training to be peformed incrementally, With the incremental formulation of an inverse matrix in optimization problem, current information and new input data can be used for building the new inverse matrix for the estimation of the optimal bias and Lagrange multipliers, so the large scale matrix inversion operation can be avoided. Numerical examples are included which indicate the performance of proposed algorithm.
Power Comparison on the effect of bandwidth for the lack of fit test
Kim,Daehak 대구효성가톨릭대학교 1998 연구논문집 Vol.58 No.2
The appropriateness of parametric modelling of repression data may be judged by comparison with a nonparametric smoothing estimators. Modern nonparametric smoothers are abundant and can be used to test the null parametric model. In nonparametric estimation problems including nonparametric regression, bandwidth selection is crucial than any other factors, But for testing problem, it is somewhat different to estimation problem. In this paper, we investigate the effect of bandwidth for testing the null parametric model based on the power comparison. We compute the power as the function of bandwidth of the test by using the bootstrap method in order to provide insight on how the type of regression curve affects the bandwidth. Also we illustrate the use of significance trace.
A Simulation Study for the Confidence Intervals of p by Using Average Coverage Probability
Kim, Daehak,Jeong, Hyeong-Chul 한국통계학회 2000 Communications for statistical applications and me Vol.7 No.3
In this paper, various methods for finding confidence intervals for p of binomial parameter are reviewed. Also we introduce tow bootstrap confidence intervals for p. We compare the performance of bootstrap methods with other methods in terms of average coverage probability by Monte Carlo simulation. Advantages of these bootstrap methods are discussed.
Bootstrapping Time series Model based on Moving blocks
Kim,Daehak 대구효성가톨릭대학교 1997 연구논문집 Vol.55 No.2
통계학의 여러 분야 중에서도 시간의 흐름에 따라 관찰되는 자료를 분석하는 시계열분석방법은 여러 분야에서 많이 응용되고 있다. 본 논문은 시계열 자료의 분석방법 중에서 기존의 방법인 잔차를 추정하고 이 추정된 잔차를 재표본 추출하는 잔차 재표본 붓스트랩 방법(Bootstrapping residual resampling method)과 크기가 ℓ인 블록들을 재표본하여 시계열 자료를 구성하고 이를 분석하는 이동브록 붓스트랩 방법(Moving blocks bootstrap)에 대하여 그 특징을 조사하고 이 두가지 붓스트랩 방법의 차이점 그리고 유사성 등을 모의 실험을 통하여 비교하였다.
Variable Bandwidth Selection for Kernel Regression
Kim, Daehak 대구효성가톨릭대학교 자연과학연구소 1994 基礎科學硏究論集 Vol.1994 No.1
In recent years, nonparametric kernel estimation of regresion function are abundant and widely applicable to many areas of statistics. Most of modern researches concerned with the fixed global bandwidth selection which can be used in the estimation of regression function with all the same value for all x. In this paper, we propose a method for selecting locally varing bandwidth based on bootstrap method in kernel estimation of fixed design regression. Performance of proposed bandwidth selection method for finite sample case is conducted via Monte Carlo simulation study.
A Comparison of Some Approximate Confidence Intervals for he Poisson Parameter
Kim, Daehak,Jeong, Hyeong-Chul 한국통계학회 2000 Communications for statistical applications and me Vol.7 No.3
In this paper, we reviewed thirteen methods for finding confidence intervals for he mean of poisson distribution. Bootstrap confidence intervals are also introduced. Two bootstrap confidence intervals are compared with the other existing eleven confidence intervals by using Monte Carlo simulation with respect to the average coverage probability of Woodroofe and Jhun (1989).