Back to results

Texas Technology University

A Geometric Approach to Least Squares

Abstract

dc:description.abstract

A routine problem is to best fit a line to planar data. The goal of this thesis is to view linear regression in a geometric light. Lines in the plane can be identified with points on the cylinder, and fitting a line to a data set can be defined by mapping from that cylinder to R. The best-fit line would then be the minimum point for that function. The two new objectives achieved in this thesis are discovering an algorithm to compute the convex hull and outlining how lines are mapped onto the cylinder.

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Ellis, Penelope M.

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2346/106850
OAI identifier oai:identifier
oai:ttu-ir.tdl.org:2346/106850

Chain of custody

source
Harvested from
Texas Technology University
Base URL
ttu-ir.tdl.org/server/oai/request
Last updated
2026-07-24
Source record
OAI-PMH GetRecord
related terms
citation

Ellis, Penelope M.. A Geometric Approach to Least Squares. 2010. https://hdl.handle.net/2346/106850