A Novel Statistical Method for Identifying Monotonic Relationships in Noisy Plots Public Deposited

http://ir.library.oregonstate.edu/concern/undergraduate_thesis_or_projects/fq977w73w

Descriptions

Attribute NameValues
Creator
Abstract or Summary
  • As scientific technologies and techniques have improved in past decades, it has become possible to quickly collect unprecedented quantities of experimental data. In the study of gene expression, for instance, datasets comprising tens of thousands of variables and hundreds of treatments can be produced in a matter of months. When analyzing three or four hundred treatments, manual verification becomes implausible.A useful statistical method should, therefore, be well suited to identify subsets of data in which a relationship is expressed while ignoring non-informative data. This paper presents a novel statistical method for identifying monotonic relationships between all variable pairs in large data sets. Specifically, it is designed to perform two functions: 1) to determine the probability that a plot is random, and 2) if the plot does not appear random, to select the subplot which is most likely to contain the signal. The method presented in this paper uses the longest monotonic path of a plot (the largest set of points which can be connected in a single monotonic path) as an indicator of signal strength, and will appropriately be referred to as Longest Path Analysis (LPA) throughout this paper.
Resource Type
Date Available
Date Copyright
Date Issued
Degree Level
Degree Name
Advisor
Non-Academic Affiliation
Keyword
Rights Statement
Language
Replaces
Additional Information
  • description.provenance : Submitted by Wanda Crannell (brr@oregonstate.edu) on 2014-07-22T21:26:40Z No. of bitstreams: 1 Eberhart-Garah thesis With Graphs 11514.pdf: 187857 bytes, checksum: 2737dff3c390adb30b7d526ab89a98b0 (MD5)
  • description.provenance : Made available in DSpace on 2014-07-22T21:51:38Z (GMT). No. of bitstreams: 1 Eberhart-Garah thesis With Graphs 11514.pdf: 187857 bytes, checksum: 2737dff3c390adb30b7d526ab89a98b0 (MD5)
  • description.provenance : Approved for entry into archive by Patricia Black(patricia.black@oregonstate.edu) on 2014-07-22T21:51:38Z (GMT) No. of bitstreams: 1 Eberhart-Garah thesis With Graphs 11514.pdf: 187857 bytes, checksum: 2737dff3c390adb30b7d526ab89a98b0 (MD5)

Relationships

Parents:

This work has no parents.

Last modified

Downloadable Content

Download PDF

Items