http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Semiparametric Median Regression with Censored Data
김대학,심주용 한국자료분석학회 2007 Journal of the Korean Data Analysis Society Vol.9 No.5
In this paper support vector machine is used for the semiparametric median regression when the responses are subject to randomly right censoring. We divide the input vector into the linear part and the nonlinear part according to whether the functional form of its effect on the response variable is specified as linear or not. The weights formed with Kaplan-Meir estimates of censoring distribution are used for the proposed method. Experimental results are then presented which indicate the performance of the estimation method.
Sparse kernel classification using IRWLS procedure
김대학 한국데이터정보과학회 2009 한국데이터정보과학회지 Vol.20 No.4
Support vector classification (SVC) provides more complete description of the linear and nonlinear relationships between input vectors and classifiers. In this paper. we propose the sparse kernel classifier to solve the optimization problem of classification with a modified hinge loss function and absolute loss function, which provides the efficient computation and the sparsity. We also introduce the generalized cross validation function to select the hyper-parameters which affects the classification performance of the proposed method. Experimental results are then presented which illustrate the performance of the proposed procedure for classification.
이항모수의 신뢰구간추정량에 대한 실제포함확률에 관한 연구
김대학,Kim, Dae-Hak 한국데이터정보과학회 2010 한국데이터정보과학회지 Vol.21 No.4
본 연구는 이항분포의 성공의 확률에 대한 신뢰구간추정량들을 비교분석하고자 한다. 일반적으로 대표본의 경우에 적용되는 잘 알려진 신뢰구간추정량과 소표본의 경우에도 적용될 수 있는 정확신뢰구간, 그리고 포아송 분포를 이용하여 구한 신뢰구간추정량과 연속성의 수정을 고려한 추정량들을 소 표본의 모의실험을 통하여 실제포함확률의 측면에서 비교하였다. In this paper, various methods for finding confidence intervals for the p of binomial parameter are reviewed. We compare the performance of several confidence interval estimates in terms of actual coverage probability by small sample Monte Carlo simulation.
김대학,Kim, Dae-Hak 한국데이터정보과학회 2010 한국데이터정보과학회지 Vol.21 No.6
본 연구는 질병자료나 사망자수 등과 관련된 자료의 분석에서 가장 많이 사용되는 초기하분포의 모수, 즉 성공의 확률에 대한 신뢰구간추정에 대하여 설펴보았다. 초기하분포의 성공의 확률에 대한 신뢰구간은 일반적으로 잘 알려져 있지 않으나 그 응용성과 활용성의 측면에서 신뢰구간의 추정은 상당히 중요하다. 본 논문에서는 초기하분포의 성공의 확률에 대한 정확신뢰구간을 소개하고 여러 가지 모집단의 크기와 표본수에 대하여, 그리고 몇가지 실현값에 대한 신뢰구간을 유도하고 소표본의 경우에 모의실험을 통하여 실제 포함확률의 측면에서 살펴보았다. In this paper, exact confidence interval of hyper-geometric parameter, that is the probability of success p in the population is discussed. Usually, binomial distribution is a well known discrete distribution with abundant usage. Hypergeometric distribution frequently replaces a binomial distribution when it is desirable to make allowance for the finiteness of the population size. For example, an application of the hypergeometric distribution arises in describing a probability model for the number of children attacked by an infectious disease, when a fixed number of them are exposed to it. Exact confidence interval estimation of hypergeometric parameter is reviewed. We consider the performance of exact confidence interval estimates of hypergeometric parameter in terms of actual coverage probability by small sample Monte Carlo simulation.
A comparison on the differential Entropy
김대학 한국데이터정보과학회 2005 한국데이터정보과학회지 Vol.16 No.3
Entropy is the basic concept of information theory. It is well defined for random varibles with known probability density function(pdf). For given data with unknown pdf, entropy should be estimated. Usually, estimation of entropy is based on the approximations. In this paper, we consider a kernel based approximation and compare it to the cumulant approximation method for several distributions. Monte carlo simulation for various sample size is conducted.
A Comparison on the Empirical Power of Some Normality Tests
김대학,엄준혁,정형철 한국데이터정보과학회 2006 한국데이터정보과학회지 Vol.17 No.1
In many cases, we frequently get a desired information based on the appropriate statistical analysis of collected data sets. Lots of statistical theory rely on the assumption of the normality of the data. In this paper, we compare the empirical power of some normality tests including sample entropy quantity. Monte carlo simulation is conducted for the calculation of empirical power of considered normality tests by varying sample sizes for various distributions.
김대학,이기락 한국데이터정보과학회 2005 한국데이터정보과학회지 Vol.16 No.4
We often extract a new feature from the original features for the purpose of reducing the dimensions of feature space and better classification. In this paper, we show feature extraction method based on independent component analysis can be used for classification. Entropy and mutual information are used for the selection of ordered features. Performance of classification based on independent component analysis is compared with principal component analysis for three real data sets.