• Communications for Statistical Applications and Methods
• /
• v.27 no.4
• /
• pp.397-412
• /
• 2020

In this paper, a notion of data depth is used to propose nonparametric multivariate two sample tests for difference between scale parameters. Data depth can be used to measure the centrality or outlying-ness of the multivariate data point relative to data cloud. A difference in the scale parameters indicates the difference in the depth values of a multivariate data point. By observing this fact on a depth vs depth plot (DD-plot), we propose nonparametric multivariate two sample tests for scale parameters of multivariate distributions. The p-values of these proposed tests are obtained by using Fisher's permutation approach. The power performance of these proposed tests has been reported for few symmetric and skewed multivariate distributions with the existing tests. Illustration with real-life data is also provided.

• Journal of the Korean Statistical Society
• /
• v.31 no.4
• /
• pp.415-431
• /
• 2002

A unified framework for statistical inference for the mean of the normal distribution to derive point estimates, confidence intervals and statistical tests is proposed. This optimal design is justified after investigating the basic information and requirements that are possible and impossible to control when specifying practical and statistical requirements. Point estimation is only credible when viewed in the larger context of interval estimation, since the information required for optimal point estimation is unspecifiable. Triple sampling is proposed and justified as a reasonable sampling vehicle to achieve the specifiable requirements within the unified framework.

• Communications for Statistical Applications and Methods
• /
• v.17 no.4
• /
• pp.493-505
• /
• 2010

Developing a test for independence of random variables X and Y against the alternative has an important role in statistical inference. Kochar and Gupta (1987) proposed a class of tests in view of Block and Basu (1974) model and compared the powers for sample sizes n = 8, 12. In this paper, we evaluate Kochar and Gupta (1987) class of tests for testing independence against quadrant dependence in absolutely continuous bivariate Farlie-Gambel-Morgenstern distribution, via a simulation study for sample sizes n = 6, 8, 10, 12, 16 and 20. Furthermore, we compare the power of the tests with that proposed by G$\ddot{u}$uven and Kotz (2008) based on the asymptotic distribution of the test statistics.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2003.11a
• /
• pp.74-78
• /
• 2003

This paper shows the dosimetric uncertainties of electromagnetic field at in-vivo and in-vitro experiments. For more accurate consequences of these researches, we have tried to find out any correlations among output power, power density and specific absorption rate(SAR) with the results of in-vivo, in-vitro tests and SAR reports of cellular phone and PDA. In the case of in-vivo tests, the power density has close statistical correlations with SAR value and in the event of in-vitro tests, the output power has considerable statistical correlations with SAR containing duty factor. On the other hand, we found that both power density and output power don't have any close correlations with SAR. And, we obtained fitted regression form among frequency, power density and SAR containing duty factor through multiple linear regression analysis.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.1
• /
• pp.34-34
• /
• 2014

We investigate the constraints on the matter density ${\Omega}m$ and the cosmological constant ${\Omega}{\Lambda}$ using the gravitational lensed QSO (Quasi Stellar Object) systems from the Sloan Digital Sky Survey (SDSS) by analyzing the distribution of image separation. The main sample consists of 16 QSO lens systems with measured source and lens redshifts. We use a lensing probability that is simply defined by the gaussian distribution. We perform the curvature test and the constraints on the cosmological parameters as the statistical tests. The statistical tests have considered well-defined selection effects and adopt parameter of velocity dispersion function. We also applied the same analysis to Monte-Carlo generated mock gravitational lens samples to assess the accuracy and limit of our approach. As the results of these statistical tests, we find that only the excessively positively curved universe (${\Omega}m+{\Omega}{\Lambda}$ > 1) are rejected at 95% confidence level. However, if the informations of the galaxy as play a lens are measured accurately, we confirm that the gravitational lensing statistics would be the most powerful tool.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.1021-1024
• /
• 2000

In this paper we describe a nonparametric Wilcoxon test and a parametric Z test based on statistical hypothesis testing for the detection of edges. We use the threshold determined by specifying significance level $\alpha$, while Bovik, Huang and Munson[4] consider the range of possible values of test statistics for the threshold. From the experimental results of edge detection, the Z method performs sensitively to the noisy image, while the Wilcoxon method is robust over both noisy nd noise-free images. Comparison with our statistical tests and Sobel operator shows that our tests perform more effectively in both noisy and noise-free images.

• Communications for Statistical Applications and Methods
• /
• v.22 no.3
• /
• pp.233-239
• /
• 2015

This study compares two generalized Lindley distributions and assesses consistency between theoretical and analytical results. Data (complete and censored) assumed to follow the Lindley distribution are generated and analyzed using two generalized Lindley distributions, and maximum likelihood estimates of parameters from the generalized distributions are obtained. Size and power of tests of hypotheses on the parameters are assessed drawing on asymptotic properties of the maximum likelihood estimates. Results suggest that whereas size of some of the tests of hypotheses based on the considered generalized distributions are essentially ${\alpha}$-level, some are possibly not; power of tests of hypotheses on the Lindley distribution parameter from the two distributions differs.

• Communications for Statistical Applications and Methods
• /
• v.14 no.3
• /
• pp.609-621
• /
• 2007

Diagnostic test results, which are approximately normal with a few number of outliers, but non-normal probability distribution, are frequently observed in practice. In the evaluation of two diagnostic tests, Greenhouse and Mantel (1950) proposed a parametric test under the assumption of normality but this test is inappropriate for the above non-normal case. In this paper, we propose a computationally simple nonparametric test that is based on quantile estimators of mean and standard deviation, instead of the moment-based mean and standard deviation as in some parametric tests. Parametric and nonparametric tests are compared with simulations under the assumption of, respectively, normality and non-normality, and under various combinations of the probability distributions for the normal and diseased groups.

• Communications for Statistical Applications and Methods
• /
• v.30 no.1
• /
• pp.109-117
• /
• 2023

Intuitively, a test with more assumptions has greater power than a test with fewer assumptions. This kind of examples are abundant in the nonparametric tests vs corresponding parametric ones. In general, the nonparametric tests are less efficient in terms of asymptotic relative efficiency (ARE) compared to corresponding parametric tests (Daniel, 1990). However, this is not always true. To test equal means under independent normal samples, the usual test involves using the t-distribution with the pooled estimator of the common variance. Adding the assumption of equal sample size, we may derive another test. In this case, two tests using more assumptions were performed for univariate (multivariate) cases. For these examples, it was found that the power function of a test with more assumptions is less than or equal to that of a test with fewer assumptions. This finding can be used as an expository example in master's mathematical statistics courses.

• Smart Structures and Systems
• /
• v.7 no.1
• /
• pp.15-25
• /
• 2011

The mechanical properties obtained from mechanical tests, such as tensile, buckling, impact and fatigue tests, are largely applied to several materials and are used today for preliminary studies for the investigation of a desired element in a structure and prediction of its behavior in use. This contribution focus on two widely used different tests: tensile and fatigue tests. Small PZT (Lead Titanate Zirconate) patches are bonded on the surface of test samples for impedance-based health monitoring purposes. Together with these two tests, the electromechanical impedance technique was performed by using aluminum test samples similar to those used in the aeronautical industry. The results obtained both from tensile and fatigue tests were compared with the impedance signatures. Finally, statistical meta-models were built to investigate the possibility of determining the state of the structure from the impedance signatures.