• Journal of the Korean Data and Information Science Society
• /
• v.9 no.2
• /
• pp.189-202
• /
• 1998

The goat of this paper is to study on variance estimation - Rice variance estimation, Gasser, Sroka and Jennen-Steinmetz's varince estimation - in smoothing goodness-of-fit tests. The comparisons of powers on test statistics are conducted by the change of variance, the number of oscillations, the amplitude of the alternative sample distribution.

• Communications for Statistical Applications and Methods
• /
• v.24 no.3
• /
• pp.303-315
• /
• 2017

Yoo (2015, Statistics and Probability Letters, 99, 109-113) derives theoretical results in an optimal sufficient dimension reduction with singular inner-product matrix. The results are promising, but Yoo (2015) only presents one simulation study. So, an evaluation of its practical usefulness is necessary based on numerical studies. This paper studies the asymptotic behaviors of Yoo (2015) through various simulation models and presents a real data example that focuses on ordinary least squares. Intensive numerical studies show that the $x^2$ test by Yoo (2015) outperforms the existing optimal sufficient dimension reduction method. The basis estimation by the former can be theoretically sub-optimal; however, there are no notable differences from that by the latter. This investigation confirms the practical usefulness of Yoo (2015).

• Journal of the Korean Data and Information Science Society
• /
• v.10 no.2
• /
• pp.429-436
• /
• 1999

We suggest test statistics for normality using Q-Q plot and P-P plot and obtain empirical quantities of these statistics. Also the power comparison with Shapiro-Wilk's W is conducted by Monte Carlo study.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.7 no.2
• /
• pp.55-58
• /
• 1982

An useful method for evaluating the threshold of the optimum detector can be used if sufficient statistics exists. This was done by giving examples. The design of the suboptimum detector can be carried out by finding the threshold of the appropriate test statistics. The results are conformed with the existing theory and the method given above is applicable in extensive area of designing the detectors.

• Communications for Statistical Applications and Methods
• /
• v.22 no.4
• /
• pp.361-375
• /
• 2015

A measure of skewness and kurtosis is proposed to test multivariate normality. It is based on an empirical standardization using the scaled residuals of the observations. First, we consider the statistics that take the skewness or the kurtosis for each coordinate of the scaled residuals. The null distributions of the statistics converge very slowly to the asymptotic distributions; therefore, we apply a transformation of the skewness or the kurtosis to univariate normality for each coordinate. Size and power are investigated through simulation; consequently, the null distributions of the statistics from the transformed ones are quite well approximated to asymptotic distributions. A simulation study also shows that the combined statistics of skewness and kurtosis have moderate sensitivity of all alternatives under study, and they might be candidates for an omnibus test.

• Communications for Statistical Applications and Methods
• /
• v.26 no.1
• /
• pp.57-67
• /
• 2019

Income distribution is a major concern in economic theory. In regional economics, it is often of interest to compare income distributions in different regions. Traditional methods often compare the income inequality of different regions by assuming parametric forms of the income distributions, or using summary statistics like the Gini coefficient. In this paper, we propose a nonparametric procedure to test for heterogeneity in income distributions among different regions, and a K-means clustering procedure for clustering income distributions based on energy distance. In simulation studies, it is shown that the energy distance based method has competitive results with other common methods in hypothesis testing, and the energy distance based clustering method performs well in the clustering problem. The proposed approaches are applied in analyzing data from China Health and Nutrition Survey 2011. The results indicate that there are significant differences among income distributions of the 12 provinces in the dataset. After applying a 4-means clustering algorithm, we obtained the clustering results of the income distributions in the 12 provinces.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.05a
• /
• pp.171-174
• /
• 2002

In this paper, diallel crosses block designs for control versus test comparisons among the lines are proposed. These designs are constructed by using partially balanced incomplete block designs with C-properties. Also, the efficiencies of the diallel crosses block designs obtained through this method are tabulated for number of lines 24 or less.

• Journal of Korean Society for Quality Management
• /
• v.42 no.2
• /
• pp.209-219
• /
• 2014

Purpose: The purpose of this study is to develop the statistical test for process capability index $C_p$ based on mixture normal process. Methods: This study uses Bootstrap method to calculate the approximate P-value for various simulation conditions under mixture normal process. Results: This study indicates that our proposed method is effective way to test for process capability index $C_p$ based on mixture normal process. Conclusion: This study finds out that statistical test for process capability index $C_p$ based on mixture normal process is useful for real application.

• The Korean Journal of Applied Statistics
• /
• v.25 no.3
• /
• pp.423-434
• /
• 2012

A two-sample test for equality of survival distribution is one of the important issues in survival analysis, especially for clinical and epidemiological research. With interval censored data, some testing methods have been developed. This study introduces some testing methods and compares them under various situations through simulation study. Based on simulation result, it provides some useful information on choosing the most appropriate testing method in a given situation.

• Communications for Statistical Applications and Methods
• /
• v.17 no.1
• /
• pp.89-98
• /
• 2010

In this paper, a two-stage group sampling plan based on the time truncated life test is proposed for the Weibull distribution. The design parameters such as the number of groups and the acceptance number in each stage are determined by satisfying the producer's and consumer's risks simultaneously when the group size and the test duration are specified. The acceptable reliability level is expressed by the ratio of the true mean life to the specified life. It was demonstrated from the comparison with single-stage group sampling plans that the proposed plan can reduce the average sample number or improve the operating characteristics.