• Journal of the Korean Data and Information Science Society
• /
• v.25 no.5
• /
• pp.971-986
• /
• 2014

When thousands of hypotheses are tested simultaneously, the probability of rejecting any true hypotheses increases, and large multiplicity problems are generated. To solve these problems, researchers have proposed different approaches to multiple testing methods, considering family-wise error rate (FWER), false discovery rate (FDR) or false nondiscovery rate (FNR) as a type I error and some test statistics. In this article, we discuss Bonferroni (1960), Holm (1979), Benjamini and Hochberg (1995) and Benjamini and Yekutieli (2001) procedures based on T statistics, modified T statistics or local-pooled-error (LPE) statistics. We also consider Sun and Cai (2007) procedure based on Z statistics. These procedures are compared in the simulation and applied to Arabidopsis microarray gene expression data to identify differentially expressed genes.

• The Korean Journal of Applied Statistics
• /
• v.27 no.7
• /
• pp.1197-1205
• /
• 2014

Typical clinical dose development studies consist of the comparison of several doses of a drug with a placebo. The primary interest is to find therapeutic window that satisfying both efficacy and safety. In this paper, we propose nonparametric method for identifying effective and safe doses in linear placement using score function. The Monte Carlo simulation is adapted to estimate the power and the family-wise error rate(FWE) of proposed procedure are compared with previous methods.

• Communications for Statistical Applications and Methods
• /
• v.19 no.1
• /
• pp.33-45
• /
• 2012

The primary interest of drug development studies is to estimate the smallest dose that shows a significant difference from the zero-dose control. The smallest dose is called the Minimum Effective dose(MED). In this paper, we suggest a nonparametric procedure to simultaneously find the MED of each group based on placements. The Monte Carlo simulation is adapted to estimate the power and the family-wise error rate(FWE) of the new procedures with those of discussed nonparametric tests to find MED.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.599-610
• /
• 2018

The method of Mack and Skillings (Technometrics, 23, 171-177, 1981) is a nonparametric multiple comparison method in a randomized block design with replications. This method is likely to result in loss of information because each block is ranked using the average of observations instead of repeated observations. In this paper, we proposed a new nonparametric multiple comparison method in the randomized block model with replications using an alignment method proposed by Hodges and Lehmann (The Annals of Mathematical Statistics, 33, 482-497, 1962) that extend the joint placement method proposed by Chung and Kim (Communications for Statistical Applications and Methods, 14, 551-560, 2007). In addition, Monte Carlo simulation compared the family wise error rate and power with the parametric method and the nonparametric method.

• The Korean Journal of Applied Statistics
• /
• v.30 no.6
• /
• pp.1027-1036
• /
• 2017

Multiple comparisons are required to confirm whether or not something is significant if the null hypothesis to test whether the difference between more than three treatments is rejected in a one-way layout. There are both parametric multiple comparison method Tukey (1953) and Nonparametric multiple comparison method based on Kruskal-Wallis (1952).This procedure is applied to a mixed sample of all data and then an average ranking is used for each of three or more treatments. In this paper, a new nonparametric multiple comparison procedure based on joint placements for a one-way layout as extension of the joint placements described in Chung and Kim (2007) was proposed. Monte Carlo simulation is also adapted to compare the family wise error rate (FWE) and the power of the proposed method with previous methods.

• Communications for Statistical Applications and Methods
• /
• v.19 no.6
• /
• pp.899-904
• /
• 2012

High-dimensional categorical data models with small sample sizes have not been used extensively in genomic sequences that involve count (or discrete) or purely qualitative responses. A basic task is to identify differentially expressed genes (or positions) among a number of genes. It requires an appropriate test statistics and a corresponding multiple testing procedure so that a multivariate analysis of variance should not be feasible. A family wise error rate(FWER) is not appropriate to test thousands of genes simultaneously in a multiple testing procedure. False discovery rate(FDR) is better than FWER in multiple testing problems. The data from the 2002-2003 SARS epidemic shows that a conventional FDR procedure and a proposed test statistic based on a pseudo-marginal approach with Hamming distance performs better.

• Communications for Statistical Applications and Methods
• /
• v.27 no.1
• /
• pp.141-148
• /
• 2020

We consider estimating the proportion of true null hypotheses in multiple testing problems. A traditional multiple testing rate, family-wise error rate is too conservative and old to control type I error in multiple testing setups; however, false discovery rate (FDR) has received significant attention in many research areas such as GWAS data, FMRI data, and signal processing. Identify differentially expressed genes in microarray studies involves estimating the proportion of true null hypotheses in FDR procedures. However, we need to account for unknown dependence structures among genes in microarray data in order to estimate the proportion of true null hypothesis since the genuine dependence structure of microarray data is unknown. We compare various procedures in simulation data and real microarray data. We consider a hidden Markov model for simulated data with dependency. Cai procedure (2007) and a sliding linear model procedure (2011) have a relatively smaller bias and standard errors, being more proper for estimating the proportion of true null hypotheses in simulated data under various setups. Real data analysis shows that 5 estimation procedures among 9 procedures have almost similar values of the estimated proportion of true null hypotheses in microarray data.

• Genomics & Informatics
• /
• v.14 no.4
• /
• pp.187-195
• /
• 2016

In genetic association studies with high-dimensional genomic data, multiple group testing procedures are often required in order to identify disease/trait-related genes or genetic regions, where multiple genetic sites or variants are located within the same gene or genetic region. However, statistical testing procedures based on an individual test suffer from multiple testing issues such as the control of family-wise error rate and dependent tests. Moreover, detecting only a few of genes associated with a phenotype outcome among tens of thousands of genes is of main interest in genetic association studies. In this reason regularization procedures, where a phenotype outcome regresses on all genomic markers and then regression coefficients are estimated based on a penalized likelihood, have been considered as a good alternative approach to analysis of high-dimensional genomic data. But, selection performance of regularization procedures has been rarely compared with that of statistical group testing procedures. In this article, we performed extensive simulation studies where commonly used group testing procedures such as principal component analysis, Hotelling's $T^2$ test, and permutation test are compared with group lasso (least absolute selection and shrinkage operator) in terms of true positive selection. Also, we applied all methods considered in simulation studies to identify genes associated with ovarian cancer from over 20,000 genetic sites generated from Illumina Infinium HumanMethylation27K Beadchip. We found a big discrepancy of selected genes between multiple group testing procedures and group lasso.