Show simple item record

dc.contributor.advisorMacchiavelli, Raúl E.
dc.contributor.authorTorres-Saavedra, Pedro A.
dc.description.abstractThis work presents different alternatives for modeling binary longitudinal data for continuous and categorical outcomes. Using a logistic model with random inter- cept, we propose the Percentile Curves concept, which are conditional probability curves across time representing percentiles of distribution of curves generated by the random intercept. We also analyze the density and cumulative distribution of subject-specific probabilities across time induced by the logistic model with random intercept. We apply this concept to two binary longitudinal data sets (Toenail and Garlic). Also, we expand the percentile curves concept to a logistic model with a random intercept and slope, and we propose a methodology to compute them.en_US
dc.description.abstractEn este trabajo se presentan diferentes alternativas de modelamiento de datos longitudinales para respuestas continuas y categóricas. Usando un modelo logístico con intercepto aleatorio, proponemos el concepto de Curvas Percentiles, que son curvas de probabilidades condicionales a través del tiempo que representan percentiles de una distribución de curvas generadas por el intercepto aleatorio. Asímismo, analizamos la función de densidad y la distribución acumulada de las proporciones sujeto-especificas a través del tiempo inducidas por el modelo logístico con intercepto aleatorio. Aplicamos este concepto a dos conjuntos de datos longitudinales binarios, Uñas y Ajo. Además, ampliamos el concepto de curvas percentiles a un modelo logístico con intercepto y pendiente aleatoria y proponemos una metodología para calcularlasen_US
dc.subjectPercentile curves concepten_US
dc.titlePercentile curves in binary longitudinal dataen_US
dc.rights.licenseAll rights reserveden_US
dc.rights.holder(c) 2006 Pedro A. Torres-Saavedraen_US
dc.contributor.committeeLorenzo, Edgardo
dc.contributor.committeeAcuña-Fernández, Edgar
dc.contributor.representativeFernando-Vega, José Statisticsen_US
dc.contributor.collegeCollege of Arts and Sciences - Sciencesen_US
dc.contributor.departmentDepartment of Mathematicsen_US

Files in this item


This item appears in the following Collection(s)

  • Theses & Dissertations
    Items included under this collection are theses, dissertations, and project reports submitted as a requirement for completing a graduate degree at UPR-Mayagüez.

Show simple item record

All rights reserved
Except where otherwise noted, this item's license is described as All Rights Reserved