• Communications for Statistical Applications and Methods
• /
• 제5권2호
• /
• pp.411-416
• /
• 1998

A simple nonparametric test of complete or total independence is suggested for continuous multivariate distributions. This procedure first discretizes the original variables based on their order statistics, and then tests the hypothesis of complete independence for the resulting contingency table. Under the hypothesis of independence, the chi-squared test statistic has an asymptotic chi-squared distribution. We present a simulation study to illustrate the accuracy in finite samples of the limiting distribution of the test statistic. We compare our method to another nonparametric test of complete independence via a simulation study. Finally, we apply our method to the residuals from a real data set.

• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1191-1200
• /
• 2006

A chi-squared test of multivariate normality is suggested which is mainly focused on detecting deviations from elliptical symmetry. This test uses Mahalanobis distances of observations to have some power for deviations from multivariate normality. We derive the limiting distribution of the test statistic by a conditional limit theorem. A simulation study is conducted to study the accuracy of the limiting distribution in finite samples. Finally, we compare the power of our method with those of other popular tests of multivariate normality under two non-normal distributions.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2006년도 PROCEEDINGS OF JOINT CONFERENCEOF KDISS AND KDAS
• /
• pp.203-212
• /
• 2006

A chi-squared test of multivariate normality is suggested which is mainly focused on detecting deviations from elliptical symmetry. This test uses Mahalanobis distances of observations to have some power for deviations from multivariate normality. We derive the limiting distribution of the test statistic by a conditional limit theorem. A simulation study is conducted to study the accuracy of the limiting distribution in finite samples. Finally, we compare the power of our method with those of other popular tests of multivariate normality under two non-normal distributions.

• 응용통계연구
• /
• 제23권4호
• /
• pp.777-781
• /
• 2010

Yates' continuity correction of the chi-squared test for testing the homogeneity of two binomial proportions in $2{\times}2$ contingency tables is developed to lower the value of the test statistic slightly. The effect of continuity correction is expected to decrease as the sample size increases. However, in extremely unbalanced $2{\times}2$ contingency tables, we find some cases where the effect of continuity correction is eccentric and is larger than expected. In such cases, we conclude that the chi-squared test with continuity correction should not be employed as a test statistic in both asymptotic tests and exact tests.

• Journal of the Korean Statistical Society
• /
• 제24권1호
• /
• pp.257-266
• /
• 1995

Consider a p-variate $(p \geq 4)$ normal distribution with mean $\b{\theta}$ and identity covariance matrix. For estimating $\b{\theta}$ under a quadratic loss we investigate the behavior of risks of Stein-type estimators which shrink the usual estimator toward the mean of observations. By using concavity of the function appearing in the shrinkage factor together with new expectation identities for noncentral chi-squared random variables, a characterization of estimators with nondecreasing risk is obtained.

• 유통과학연구
• /
• 제14권7호
• /
• pp.5-14
• /
• 2016

Purpose - This paper studies if types of markets have an association with several variables such as occupation, district, sales, employee, administrative district by a lessor, area in rental building and so on. Three attributes of markets can be general commercial zones, central commercial zones and traditional markets. Furthermore, we can investigate the degree of association by calculating the numerical strength and visualizing their distances on two-dimensional plane, once the association exists between them. Research design, data, and methodology - This work is performed by the 2013 report presented with Small Businessmen Promotion Institute in Korea and used by a chi-squared test and correspondence analysis by using IBM SPSS 23.0. Results - The results show that types of markets, including traditional markets, have an association with variables considered in this paper, and we can obtain the detailed associations between attributes of corresponding variables by inspecting two-dimensional plane. Conclusions - This study suggests that government authority and local autonomy can make strategies to vitalize traditional markets and to get win-win relationships among several types of markets by looking over our findings.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.375-387
• /
• 1995

Main purpose of this article is to consider the asymptotic distribution of the rank transformed F statistic for interaction in a two-way layout. Some theorems and sufficient conditions are derived to have the rank transformed F statistic converged in distribution to a chi-squared random variable with (I-1)(J-1) degrees of freedom divided by (I-1)(J-1). These results will be useful for the other theoretical studies of the rank transform procedure in experimental designs.

• 응용통계연구
• /
• 제19권3호
• /
• pp.545-554
• /
• 2006

로그선형모형에 기반을 둔 다차원 임의 분할표를 생성하는 방법을 제안하였다. 이를 위해 Lee(1997)가 제안한 선형 결합에 의한 결합분포 생성 방법을 적용하였으며, Pearson 통계량을 연관성 측도로 사용하는 것을 제안하였다. 세 변수가 서로 완전한 연관을 갖는 삼차원 결합분포를 생성할 수 있으므로 본 연구에서 제안한 방법은 사차원 이상 다차원 임의 분할표를 생성하는 문제로 확장될 수 있다.

• 응용통계연구
• /
• 제18권1호
• /
• pp.99-113
• /
• 2005

이 논문에서는 카이제곱 구형 대칭성 검정을 제안한다. 이 검정은 검정통계량과 점근적 유의확률을 쉽게 계산할 수 있는 장점이 있다. 이 통계량의 구형 대칭성 가정하의 극한 분포를 도출하고 유한표본에서 잘 부합되는지 모의실험을 통해 살펴본다. 또한 다양한 대립분포에서 기존의 구형 대칭성 검정과 검정력을 비교하는 모의실험을 수행하며 마지막으로 실제 자료 분석 예제를 제공한다.

• 산경연구논집
• /
• 제7권3호
• /
• pp.5-12
• /
• 2016

Purpose - In this paper, we consider more segmented types of markets than conventional version of ones in South Korea and explore the degree of relations between these markets and the related factors with them. In this case, ten attributes of types of markets mentioned above will be considered. To be more specific, the numerical strength is evaluated and graphical approach is expressed on two-dimensional plane, if the association exists between the considered variables. Research design, data, and methodology - This work is done by the 2013 report on the commercial building lease offered by Small Businessmen Promotion Institute (May/2013~August/2013) and exploited by statistical analyses such as correspondence analysis and a chi-squared test in IBM SPSS 23.0. Results - Findings of this paper indicate that a variable Korean market, including traditional markets, are closely connected with variables administrative district, sales and occupation instead of company, age group and business duration and the detailed associations between variables can be obtained by inspecting results of correspondence analysis. Conclusions - We can understand where the status of the Korean markets stands now through this work and also government authority and local autonomy can take advantage of these findings to enhance the revitalization of Korean markets and other markets.