GUPEA >
Faculty of Science / Naturvetenskapliga fakulteten >
Department of Mathematical Sciences / Institutionen för matematiska vetenskaper >
Doctoral Theses / Doktorsavhandlingar Institutionen för matematiska vetenskaper >

Extreme Value Analysis of Huge Datasets: Tail Estimation Methods in High-Throughput Screening and Bioinformatics


Please use this identifier to cite or link to this item: http://hdl.handle.net/2077/27833

Files in This Item:

File Description SizeFormat
gupea_2077_27833_1.pdfPhD Thesis2730KbAdobe PDF
View/Open
gupea_2077_27833_2.pdfAbstract117KbAdobe PDF
View/Open
Title: Extreme Value Analysis of Huge Datasets: Tail Estimation Methods in High-Throughput Screening and Bioinformatics
Authors: Dmitrii, Zholud
E-mail: dmitrii@zholud.com
Issue Date: 13-Oct-2011
University: Göteborgs universitet. Naturvetenskapliga fakulteten
Institution: Department of Mathematical Sciences ; Institutionen för matematiska vetenskaper
Parts of work: I. Rootz ́en, H. and Zholud, D.S. (2011). Tail estimation methods for the number of false positives in high-throughput testing. Submitted.

II. Zholud, D.S. (2011). Tail approximations for the Stu- dent t−, F−, and Welch statistics for non-normal and not necessarily i.i.d. random variables. Submitted.

III. Zholud, D.S. (2009). Extremes of the Shepp statistic for a Gaussian random walk. Extremes, 12(1):1-17.
VIEW ARTICLE


IV. Zholud, D.S. (2008). Extremes of the Shepp statistic for the Wiener process. Extremes, 11(4):339-351.
VIEW ARTICLE

Date of Defence: 2011-11-03
Disputation: Torsdagen den 3 november 2011, kl. 10.15, Hörsal Pascal, Matematiska Vetenskaper, Chalmers Tvärgata 3
Degree: Doctor of Philosophy
Publication type: Doctoral thesis
Keywords: Extreme Value Statistics, High-Throughput Screening, HTS, Bioinformatics, analysis of huge datasets, quality control, correction of theoretical p-values, comparison of pre-processing methods, SmartTail, estimation of False Discovery Rates, test power, distribution tail, high level excursions, quantile estimation, multiple testing, Student t−test, Welch statistic, small sample sizes, F−test, Wiener process, Gaussian random walk, Shepp statistic, limit theorems, exotic options.
Abstract: This thesis presents results in Extreme Value Theory with applications to High-Throughput Screening and Bioinformatics. The methods described here, however, are applicable to statistical analysis of huge datasets in general. The main results are covered in four papers. The first paper develops novel methods to handle false rejections in High-Throughput Screening experiments where testing is done at extreme significance levels, with low degrees of freedom, and when the true null distributi... more
ISBN: 978-91-628-8354-6
URI: http://hdl.handle.net/2077/27833
Appears in Collections:Doctoral Theses / Doktorsavhandlingar Institutionen för matematiska vetenskaper
Doctoral Theses from University of Gothenburg / Doktorsavhandlingar från Göteborgs universitet

This item has been viewed 131 times.

 

 

© Göteborgs universitet 2011