Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/81845
Title: Análisis de correlación moderno: ¿Qué alternativas existen para la correlación de Pearson?
Author: Pazos Ruiz, Ana Belén
Tutor: Sánchez-Pla, Alex  
Others: Universitat Oberta de Catalunya
Abstract: The Pearson correlation coefficient finds the linear dependence between two variables with a straightforward and low computational cost, but having to meet some assumptions difficult to assume. Currently, there is a high number of coefficients that identify the association between variables, among them are the coefficients CorGC, RDC, dCor or MIC, the latter named as the correlation of s. XXI. In this work have been analysed these coefficients to find the most complete, distinguishing which detect more types of associations with a low computational cost and fulfilling the seven fundamental properties proposed by Rényi. They were tested with eight different types of associations (pseudo-random, linear, quadratic, cubic, exponential, sinusoidal, step and circle), as well as a gene database in which it was necessary to know which genes were expressed significantly. It also provides a straightforward application to calculate four of them.
Keywords: correlation analysis
independence
non-linear association
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 5-Jun-2018
Publication license: http://creativecommons.org/licenses/by-nc-sa/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
Memoria_AnaBelénPazosRuiz.docx1,69 MBMicrosoft Word XMLView/Open
apazosrTFM0618memoria.pdfMemoria del TFM1,23 MBAdobe PDFThumbnail
View/Open