Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 4 of 4
Full-Text Articles in Physical Sciences and Mathematics
Combining Functions And The Closure Principle For Performing Follow-Up Tests In Functional Analysis Of Variance, Olga A. Vsevolozhskaya, Mark C. Greenwood, G. J. Bellante, S. L. Powell, R. L. Lawrence, K. S. Repasky
Combining Functions And The Closure Principle For Performing Follow-Up Tests In Functional Analysis Of Variance, Olga A. Vsevolozhskaya, Mark C. Greenwood, G. J. Bellante, S. L. Powell, R. L. Lawrence, K. S. Repasky
Olga A. Vsevolozhskaya
Functional analysis of variance involves testing for differences in functional means across kk groups in nn functional responses. If a significant overall difference in the mean curves is detected, one may want to identify the location of these differences. Cox and Lee (2008) proposed performing a point-wise test and applying the Westfall–Young multiple comparison correction. We propose an alternative procedure for identifying regions of significant difference in the functional domain. Our procedure is based on a region-wise test and application of a combining function along with the closure multiplicity adjustment principle. We give an explicit formulation of how to implement …
Use Of P-Values To Evaluate The Probability Of A Genuine Finding In Large-Scale Genetic Association Studies, Olga A. Vsevolozhskaya, Qing Lu, Chia-Ling Kuo, Dmitri V. Zaykin
Use Of P-Values To Evaluate The Probability Of A Genuine Finding In Large-Scale Genetic Association Studies, Olga A. Vsevolozhskaya, Qing Lu, Chia-Ling Kuo, Dmitri V. Zaykin
Olga A. Vsevolozhskaya
To claim the existence of an association in modern genome-wide association studies (GWAS), a nominal P-value has to exceed a stringent Bonferroni-adjusted significance level. Despite strictness of the correction, a significant P-value does not indicate high probability that the claimed association is genuine. A simple Bayesian solution -- the False Positive Report Probability (FPRP) -- was previously proposed to convert the observed P-value to the corresponding probability of no true association. Although the FPRP solution is highly popular, it does not reflect probability that a particular finding is false. Here, we offer a simple POFIG method -- a Probability that …
Association Studies For Sequencing Data With Functional Analysis Of Variance, Olga A. Vsevolozhskaya, Mark C. Greenwood, Changshuai Wei, Qing Lu
Association Studies For Sequencing Data With Functional Analysis Of Variance, Olga A. Vsevolozhskaya, Mark C. Greenwood, Changshuai Wei, Qing Lu
Olga A. Vsevolozhskaya
The rapid development of next generation sequencing technologies and accompanying reduction in cost produce an increasing number of single nucleotide polymorphisms (SNPs) that can be identified across the genome. Analyzing high-dimensional genomic data is a challenge and requires development of new statistical methods. We propose to use the functional analysis of variance (FANOVA) to perform inference for sequencing data. FANOVA is used to test for differences in functional means of k groups over time. We suggest using FANOVA to test for a significant difference among SNPs between levels of a phenotype, such as the presence or absence of a disease. …
Resampling-Based Multiple Comparison Procedure With Application To Point-Wise Testing With Functional Data, Olga A. Vsevolozhskaya, Mark C. Greenwood
Resampling-Based Multiple Comparison Procedure With Application To Point-Wise Testing With Functional Data, Olga A. Vsevolozhskaya, Mark C. Greenwood
Olga A. Vsevolozhskaya
No abstract provided.