Bibliographic citations
Baldeon, D., (2024). Regresión cuantílica binaria: un enfoque bayesiano basado en la distribución asimétrica de Laplace [Pontificia Universidad Católica del Perú]. http://hdl.handle.net/20.500.12404/27108
Baldeon, D., Regresión cuantílica binaria: un enfoque bayesiano basado en la distribución asimétrica de Laplace []. PE: Pontificia Universidad Católica del Perú; 2024. http://hdl.handle.net/20.500.12404/27108
@mastersthesis{renati/536089,
title = "Regresión cuantílica binaria: un enfoque bayesiano basado en la distribución asimétrica de Laplace",
author = "Baldeon Molleda, Dante Reynaldo",
publisher = "Pontificia Universidad Católica del Perú",
year = "2024"
}
Quantile regression is a statistical technique that allows for the analysis of relationships between variables across different quantiles of the response variable’s distribution. However, its application to binary response variables can be counterintuitive, as the traditional definition of quantiles is conceptualized for continuous variables and does not have a direct interpretation in a binary variable. Although a binary response variable only takes two values and does not allow for a traditional definition of quantiles, it is possible to extend quantile regression to model the quantiles of the latent variable underlying the binary response variable. This latent variable is continuous and enables the application of quantile regression in contexts where the response variable is binary. In this study, we adopt a Bayesian approach to binary quantile regression based on the Asymmetric Laplace Distribution (ALD); we will apply the model to a dataset comprising discarded COVID-19 test results in oncology patients and estimate the regression coefficients using the bayesQR package developed in R.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.