South Dakota State University
U-Statistics for Characterizing Forensic Sufficiency Studies
Abstract
dc:description.abstractOne of the main metrics for deciding if a given forensic modality is useful across a broad spectrum of cases, within a given population, is the Random Match Probability (RMP), or the corresponding discriminating power. Traditionally, the RMP of a given modality is studied by comparing full `templates' and estimating the rate at which pairs of templates 'match' in a given population. This strategy leads to a natural U-statistic of degree two. However, in questioned document examination, the RMP is studied as a function of the amount of handwriting contained in the two documents being compared; turning the U-statistic into a U-process. This work is focused on providing background on forensic sufficiency studies, RMP, and the U-processes that naturally arise in this class of problems.
Degree
thesis:*- Name thesis:degree_name
- Master of Science (MS)
- Level thesis:degree_level
- Thesis - Open Access
- Discipline thesis:degree_discipline
- Mathematics and Statistics
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Fuglsby, Cami
- Contributors dc:contributor
-
- Christopher P. Saunders
Subjects
dc:subject × 6Rights
dc:rights- Language dc:language
- en
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://openprairie.sdstate.edu/etd/1715