• The Korean Journal of Applied Statistics
• /
• v.1 no.2
• /
• pp.66-81
• /
• 1987

A new nonparametric sequential testing procedure is proposed in the paper. Sequential observations are divided into equally sized groups and a nonparametric statistic, which is appropriate for testing the given hypotheses, is obtained from each group. Then Wald's sequential test is applied for the case where the log probability ratio statistic is replaced by the nonparametric statistic. The properties of such test are evaluated approximately by the Wiener process.

• Communications for Statistical Applications and Methods
• /
• v.15 no.2
• /
• pp.283-292
• /
• 2008

Smoothing parameter or bandwidth plays a key role in nonparametric classification based on kernel density estimation. We consider choosing smoothing parameter in nonparametric classification, which optimize the Bayes risk. Hall and Kang (2005) clarified the theoretical properties of smoothing parameter in terms of minimizing Bayes risk and derived the optimal order of it. Bootstrap method was used in their exploring numerical properties. We compare cross-validation and bootstrap method numerically in terms of optimal order of bandwidth. Effects on misclassification rate are also examined. We confirm that bootstrap method is superior to cross-validation in both cases.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1171-1180
• /
• 2010

It is well known that the asymptotic convergence rates of nonparametric regression estimator gets worse as the dimension of covariates gets larger. One possible way to overcome this problem is reducing the dimension of covariates by using single index models. Two coefficient estimation methods in single index models are introduced. One is semiparametric least square estimation method, which tries to find approximate solution by using iterative computation. The other one is weighted average derivative estimation method, which is non-iterative method. Both of these methods offer the parametric convergence rate to normal distribution. However, practical comparison of these two methods has not been done yet. In this article, we compare these methods by examining the variances of estimators in various models.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.599-610
• /
• 2018

The method of Mack and Skillings (Technometrics, 23, 171-177, 1981) is a nonparametric multiple comparison method in a randomized block design with replications. This method is likely to result in loss of information because each block is ranked using the average of observations instead of repeated observations. In this paper, we proposed a new nonparametric multiple comparison method in the randomized block model with replications using an alignment method proposed by Hodges and Lehmann (The Annals of Mathematical Statistics, 33, 482-497, 1962) that extend the joint placement method proposed by Chung and Kim (Communications for Statistical Applications and Methods, 14, 551-560, 2007). In addition, Monte Carlo simulation compared the family wise error rate and power with the parametric method and the nonparametric method.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.461-473
• /
• 2014

We consider condence intervals for high quantiles of heavy-tailed distribution. The asymptotic condence intervals based on the limiting distribution of estimators are considered together with bootstrap condence intervals. We can also apply a non-parametric, parametric and semi-parametric approach to each of these two kinds of condence intervals. We considered 11 condence intervals and compared their performance in actual coverage probability and the length of condence intervals. Simulation study shows that two condence intervals (the semi-parametric asymptotic condence interval and the semi-parametric bootstrap condence interval using pivotal quantity) are relatively more stable under the criterion of actual coverage probability.

• The Korean Journal of Applied Statistics
• /
• v.11 no.1
• /
• pp.193-204
• /
• 1998

It is quite that we have to estimate the derivative of the regression function. The Bezier curve, rarely known to statisticians, is very popular in computer graphics area. In this paper, we use nonparametric method via the Bezier curve, and apply this method to real data set. This method seems to be very easy to compute and can be easily applied to other smoothing techniques.

• Proceedings of the Korea Database Society Conference
• /
• 1998.09a
• /
• pp.584-594
• /
• 1998

선형회귀분석은 가장 널리 사용되는 데이터 분석기법이지만 독립변수와 종속변수간의 관계가 선형이라고 가정하기 때문에 문제점을 가지고 있다. 비모수 회귀분석(Nonparametric Regression)은 선형회귀분석의 문제점을 극복할 수 있는 방법으로 변수간의 관계의 형태를 미리 가정하지 않고 데이터에 의해 결정하는 방법이다. 본 연구에서는 유전자 알고리즘을 비모수 회귀분석법 중의 하나인 Regressoin Splines에 적용하였다. 인위적 데이터를 이용한 평가 결과 유전자 알고리즘은 다양한 상황에서 매우 우수한 것으로 나타났다.

• Journal of Labour Economics
• /
• v.29 no.2
• /
• pp.145-169
• /
• 2006

This study analyze the effect on wage, labor mobility by using Craft II Class certification out of National skill certification. In this article, we used the parametric and nonparametric method. In the former we used IV that the fraction of certification by occupation by firm scale to solve the selection problem. In the latter, it's used matching method and kernel regression. The paper shows that certification effect on wage has about 5.1~9.9%. The result of analysis between certification and labor mobility indicates better certification effects on long term tenure to the same firm than certification effects on wage from labor mobility. Also, we knew that the employee which have no certification relative is difficult to be established in the same workplace.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.193-196
• /
• 2007

본 논문은 비모수적 클러스터링 기법을 이용하여 다양한 조명에 노출된 의상들의 색상 유사성을 안정적으로 판단하는 방법을 제안한다. 색상 유사성 판별을 위하여 기존에 대표적으로 사용되어왔던 히스토그램 인터섹션이나 누적 히스토그램 방법은 조명 변화에 민감하게 반응하여, 동일한 의상 색상이라 할지라도 서로 다른 조명환경에서는 서로 상이한 색상 판별 결과를 나타낸다. 본 논문에서는 조명에 의한 영향을 줄이고, 색상 자체의 분포 특성을 분석하기 위하여 조명조건의 변화에도 일관된 특성을 유지하는 색도와 채도 컬러 성분에 대한 분포 특성을 비모수적 클러스터링 기법을 적용하여 분석한다. 실험 결과 제안기법은 동일한 의상 쌍과 상이한 의상 쌍에 대하여 구분을 지을 수 있는 양자화의 특성이 뚜렷하게 표현되었다.

• The Korean Journal of Applied Statistics
• /
• v.30 no.2
• /
• pp.281-290
• /
• 2017

Mack and Skillings (1980) proposed a nonparametric method in a randomized block design with replications. This method employs the mean of observations instead of each observation. However, it has the inherent disadvantage that there may be a loss of information. In this paper, we proposed a nonparametric method that employees an aligned method and linear placement statistics to supplement its weakness. A Monte-Carlo study is performed to compare the power of the proposed method with previous methods.