• Journal of the Korean Statistical Society
• /
• v.27 no.2
• /
• pp.189-195
• /
• 1998

In this paper, we are concerned with the problem of testing the randomness of the coefficients in a first order autoregressive model. A consistent test based on prediction error is suggested. It is shown that under the null hypothesis, the test statistic is asymptotically normal.

• Communications for Statistical Applications and Methods
• /
• v.21 no.4
• /
• pp.309-316
• /
• 2014

Testing normality is very important because the most common assumption is normality in statistical analysis. We propose a new plot and test statistic to goodness-of-fit test for normality based on the generalized Lorenz curve. We compare the new plot with the Q-Q plot. We also compare the new test statistic with the Kolmogorov-Smirnov (KS), Cramer-von Mises (CVM), Anderson-Darling (AD), Shapiro-Francia (SF), and Shapiro-Wilks (W) test statistic in terms of the power of the test through by Monte Carlo method. As a result, new plot is clearly classified normality and non-normality than Q-Q plot; in addition, the new test statistic is more powerful than the other test statistics for asymmetrical distribution. We check the proposed test statistic and plot using Hodgkin's disease data.

• The Korean Journal of Applied Statistics
• /
• v.30 no.2
• /
• pp.233-241
• /
• 2017

Interval censored data frequently occur in observation studies where the subject is followed periodically. In this paper, our interest is to suggest a test statistic to compare the CIF of two groups with interval censored failure time data in the presence of competing risks. Gray (1988) suggested a test statistic for right censored data that motivated a well-known Fine and Gray's subdistribution hazard model. A multiple imputation technique is adopted to adopt Gray's test statistic to interval censored data. The powers and sizes of the suggested method are investigated through diverse simulation schemes. The main merit of the suggested method is its simplicity to implement with existing software for right censored data. The method is illustrated by analyzing Bangkok's HIV cohort dataset.

• Journal of the Korean Statistical Society
• /
• v.13 no.1
• /
• pp.32-41
• /
• 1984

An adaptive distribution-free test is proposed for testing the equality of k independent distributions against unrestricted alternatives. In this paper, several rank-sum test statistics are considered as teh components of the adaptive one. The emprical powers of the adaptive testing procedure are compared to those of the classical F test and the component tests through a Monte Carlo study. The results show that the adaptive test has good power properties over a wide class of underlying distributions.

• Journal of the Korean Statistical Society
• /
• v.15 no.1
• /
• pp.9-25
• /
• 1986

In this paper we propose a distribution-free rank test for ordered alternatives in a randomized block design and investigate the properties of the proposed test. The proposed test is an extension of the Page test to allow replications in each cell. Some asymptotic properties including ARE's are investigated. A small sample Monte Carlo study was performed to compare the powers of the test considered in this paper for small samples. The results show that our proposed test is robust and efficient in the case of equally-spaced treatment effects.

• Journal of applied mathematics & informatics
• /
• v.3 no.2
• /
• pp.235-244
• /
• 1996

Ranked-set sampling is useful when measurements are destructive or costly to obtain but ranking of the observations is rel-atively easy. The Wilcoxon signed rank test statistic based on the ranked-set sample is considered. We compared the asymptotic relative efficiencies of the RSS Wilcoxon signed rank test statistics with respect to the SRS Wilcoxon signed rank test statistic and the RSS sign test statistic. Throughout the ARE's the proposed test statistic is superior to the SRS Wilcoxxon signed rank test statistic and the RSS sign test statistic.

• Journal of the Korean Statistical Society
• /
• v.29 no.3
• /
• pp.351-360
• /
• 2000

This paper considers the problem of testing a variance change in infinite order autoregressive models. A cusum of squares test based on the residuals from an AR(q) model is constructed analogous to Inclan and Tiao (1994)'s test statistic, where q is a sequence of positive integers diverging to $\infty$. It is shown that under regularity conditions the limiting distribution of the test statistic is the sup of a standard Brownian bridge. Simulation results are given to illustrate the performance of the test.

• Journal of the Korean Statistical Society
• /
• v.18 no.1
• /
• pp.38-45
• /
• 1989

A procedure is considered to the problem of testing whether there exist changes in location at possibly two points in a sequence of independent random variables which are successively drawn from normal population. A test statistics based on modified likelihood ratio is proposed and its asymptotic null distribution is derived through the stochastic process representation. A small sample power comparison is made by Monte Carlo method.

• Journal of Korean Society for Quality Management
• /
• v.22 no.1
• /
• pp.152-161
• /
• 1994

In this paper we develop a test for alternatives representing decreasing mean residual life. The test statistic for decreasing mean residual life, $K_{1n}$, is a modified version of Hollander and Proschan's test $V^*$ and critical constants and large sample approximation are shown to make the test readily applicable. Consistency is also shown for the tests based on $K_{1n}$. And small sample powers for four alernatives are obtained.

• Journal of the Korean Statistical Society
• /
• v.22 no.2
• /
• pp.325-337
• /
• 1993

This paper propose two test statistics which enable us to proceed the variable selection in Fisher's linear discriminant function for the case of heterogeneous discrimination with equal training sample size. Simultaneous confidence intervals associated with the test are also given. These are exact and approximate results. The latter is based upon an approximation of a linear sum of Wishart distributions with unequal scale matrices. Using simulated sampling experiments, powers of the two tests have been tabulated, and power comparisons have been made between them.