Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/146106
Title: Missing data: Imputación múltiple en bases de datos pequeñas
Author: Hernández Villena, Juan Vicente
Tutor: Perez-Alvarez, Nuria  
Others: Ventura, Carles  
Abstract: A missing data is relevant information to the analysis, but due to different factors could not be recorded, and as consequence, is absent in any kind of dataset, including longitudinal records. If they are not taken into account, they will influence significantly the statistical power, the analysis integrity, the bias, and the quality of the results. So, it is necessary to use the correct treatment on them, based on their characteristics. This study carried out several multiple imputation alternatives by the PMM strategy, a modern treatment that has generated good results, on an HIV dataset missing data, in three scenarios according to the sample size. Once the imputations were done, the results were put to test, reproducing the analysis carried out in the original paper where the data comes from (comparison between treatments), obtaining similar results to those described. An improved workflow was described for the missings¿ previous analysis and treatment, with a longitudinal dataset, with small sample size, non-normal distribution, and with high percentages of missing data, characteristics that are not common when these methods are evaluated, but which are common in the most research fields such as biology and healthcare.
Keywords: missing data
bioinformatics
HIV
multiple imputation
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jun-2022
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 
juanvhernandezvTFM0622memoria.pdfMemoria del TFM1,24 MBAdobe PDFThumbnail
View/Open
juanvhernandezvTFM0622presentación.pdfPresentación del TFM827,77 kBAdobe PDFThumbnail
View/Open