• 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.

• The Korean Journal of Applied Statistics
• /
• v.19 no.1
• /
• pp.111-120
• /
• 2006

In a two-stage two-level balanced nested design, type I error rates for the parametric tests and the rank transformed tests for the main effects and the nested effects are in overall similar to each other. Furthermore, powers for the rank transformed statistic for the main effects and the nested effects in a two-stage two-level balanced nested design are generally superior to powers for the parametric statistic When the effect size and the sample size are increased, we can find that powers increase for the parametric statistic and the rank transformed statistic are dramatically improved. Especially for the case of the fixed effects in the asymmetric distributions such as an exponential distribution, powers for the rank transformed tests are quite high rather than powers for the parametric tests.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.8C
• /
• pp.709-712
• /
• 2010

In this paper, the two sample locally optimum rank detector is obtained in the weakly dependent noise with non-zero temporal correlation between noise observations. The test statistic of the locally optimum rank detector is derived from the Neyman-Pearson lemma suitable for the two sample observation models, where it is assumed that reference observations are available in addition to regular observations. Two-sample locally optimum rank detecter shows the same performance with the one-sample locally optimum rank detector asymptotically. The structure of the two-sample rank detector is simpler than that of the one-sample rank detector because the sign statistic is not processed separately.

• Journal of the Korean Statistical Society
• /
• v.25 no.3
• /
• pp.419-432
• /
• 1996

When interactions are not present in a three-way layout, the lim-iting null distribution of the F statistic for testing main effects when applied to the rank-score transformed data is the same as the limiting null distribution of the usual F statistic when applied to the normal data. The simulation results exhibit that the rank transform test is robust with respect to significance level and powerful for testing main effects in a three-way factorial experiment.

• Journal of the Korean Statistical Society
• /
• v.24 no.2
• /
• 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.

• Communications for Statistical Applications and Methods
• /
• v.23 no.3
• /
• pp.215-230
• /
• 2016

The powers of some tests for independence hypothesis against positive (negative) quadrant dependence in generalized Farlie-Gumbel-Morgenstern distribution are compared graphically by simulation. Some of these tests are usual linear rank tests of independence. Two other possible rank tests of independence are locally most powerful rank test and a powerful nonparametric test based on the $Cram{\acute{e}}r-von$ Mises statistic. We also evaluate the empirical power of the class of distribution-free tests proposed by Kochar and Gupta (1987) based on the asymptotic distribution of a U-statistic and the test statistic proposed by $G{\ddot{u}}ven$ and Kotz (2008) in generalized Farlie-Gumbel-Morgenstern distribution. Tests of independence are also compared for sample sizes n = 20, 30, 50, empirically. Finally, we apply two examples to illustrate the results.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1195-1205
• /
• 2014

Compared with single design, powers of rank transformed statistic for testing main and interaction effects for $2{\times}2$ factorial in $4{\times}4$ latin square design are rapidly increased as effect size and replication size are increased. In general powers of rank transformed statistic are superior without regard to the diversified effect composition and the type of error distributions as nontesting factors are few and effect size are small. Powers of rank transformed statistic show much higher level than those of parametric statistic in exponential and double exponential distributions. Further powers of rank transformed statistic are very similar with those of parametric statistic in normal and uniform distributions.

• Communications for Statistical Applications and Methods
• /
• v.5 no.1
• /
• pp.133-144
• /
• 1998

In this paper, we propose the two-sample test statistic using Wilcoxon signed rank test on ranked-set sampling(RSS) and obtain the asymptotic relative efficiencies(ARE) of the proposed test statistic with respect to Mann-Whitney-Wilcoxon statistic on simple random sampling(SRS), the Mann-Whitney-Wilcoxon statistic on RSS, sign statistic on RSS and Wilcoxon signed rank test on SRS. From the simulation works, we compare the powers of the proposed test statistic, Mann-Whitney-Wilcoxon statistic on RSS, the usual two-sample t statistic, sign statistic on RSS, where the underlying distributions are uniform, normal, double exponential, logistic and Cauchy distributions.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.1
• /
• pp.143-152
• /
• 2017

Restriction of completely randomization within a block can be handled by a split plot factorial design splitted by several plots. $3{\times}3$ split plot factorial design with two fixed main factors and one fixed block shows that powers of the rank transformed statistic for testing whole plot factorial effect and split plot factorial effect are superior to those of the parametric statistic when existing effect size is small or the remaining effect size is relatively smaller than the testing factorial effect size. Powers of the rank transformed statistic show relatively high level for exponential and double exponential distributions, whereas powers of the parametric and rank transformed statistic maintain similar level for normal and uniform distributions. Powers of the parametric and rank transformed statistic with two fixed main factors and one random block are respectively lower than those with all fixed factors. Powers of the parametric andrank transformed statistic for testing split plot factorial effect with two fixed main factors and one random block are slightly lower than those for testing whole plot factorial effect, but powers of the rank transformed statistic show comparative advantage over those of the parametric statistic.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.2
• /
• pp.245-253
• /
• 2011

Powers of rank transformed statistic for testing main effects and interaction effects for $2{\times}2$ factorial design in randomized complete block design are very superior to powers of parametric statistic without regard to the block size, composition method of effects and the type of population distributions such as exponential, double exponential, normal and uniform. $2{\times}2$ factorial design in RCBD increases error effects and decreases powers of parametric statistic which results in conservativeness. However powers of rank transformed statistic maintain relative preference. In general powers of rank transformed statistic show relative preference over those of parametric statistic with small block size and big effect size.