Facultad de Estadística
Recent Submissions
-
Modelamiento de redes sociales múltiples
In this work, we propose a statistical model to simultaneously characterize two or more social networks, where the same set of actors interact with each other. In addition to investigate ties within networks, the proposed ... -
Una aplicación de redes neuronales artificiales para el pronostico de los rendimientos de la serie de Ecopetrol
This work proposes a model based on recurrent neural networks (RNN) to forecast the performance of Ecopetrol, which is one of the most successful actions in the country. Various configurations of RNN are studied, including ... -
Brecha de género en programas de formación en Estadística en Colombia entre los años 2007-2018
One of the problems in higher education in Colombia is the existence of gender gaps, these can be observed either by career or by the region in which it is offered, it can also be affected by the level of training to which ... -
Comparación de modelos clásicos en series de tiempo y modelos bayesianos para pronosticar tres acciones colombianas en el último año
In this work, it is determined that statistical model is best for forecasting the yields of Colombian financial shares(Ecopetrol S.A (ECO), Grupo Nutresa S.A (NCH) and Banco Davivienda Pf (DVI_p), Models applied to series ... -
Modelo logístico multinomial ordinal para la caracterización del dengue en el departamento del Caquetá entre los años 2012 hasta el 2019
Dengue is a viral disease that affects the vast majority of countries in the tropics and subtropics, being transmitted by the Aedes aegypti mosquito. In the department of Caquetá, being a humid area with a large number of ... -
Procesos puntuales poisson para evaluar la distribución espacial de yacimientos arqueológicos en Colombia
The archeology and the spatial distribution of the different archaeological sites allow to understand the behavior and interaction with the territory of those who occupied it. The present document proposes the identification ... -
Algoritmos de aprendizaje supervisado utilizando datos de monitoreo de condiciones: Un estudio para el pronóstico de fallas en máquinas
This paper provides an overview of some Machine Learning and Deep Learning methods as fundamental tools in detecting potential failures of physical assets using condition monitoring techniques, for this, in the first part ... -
Propuesta metodológica para la construcción de un índice de transformación digital de los clientes en la empresa Enel Codensa con análisis factorial de datos mixto (FAMD)
Digital transformation is one of the most relevant issues in business planning currently, in an increasingly globalized world and with customers interacting more digitally, digital transformation is defi ned as the proper ... -
Acercamiento al comercio entre Colombia y Japón usando técnicas multivariadas para el período enero 2001 diciembre 2018
Japan for being one of the main economies in the world, represents a great opportunity for Colombian exporters, who historically have focused their sights on closer markets such as the US and the European. However, Asia ... -
Diseño y evaluación del efecto de los extractos provenientes de pennisetum purpureum sobre la germinación y crecimiento de la especie forestal cedrela odorata l.
This study presents the application of statistical models for the analysis of growth and germination of the species Cedrela Odorata L. It is proposed the adjustment of a linear mixed model for growth and a Poisson linear ... -
Comparacion de dos modelos para predecir la probabilidad deincumplimiento (Modelo Logit-Modelo GAM)
One of the most important needs of lending institutions is to have reliable criteria to determine who should grant the credit. In this sense, methodologies such as credit granting models should be implemented as a solution ... -
Modelo de curvas de crecimiento: Modelos lineales mixtos vs Datos funcionales
A comparison is made between the use of mixed linear models, explicitly of growth curves for the analysis of longitudinal data against an application of functional data by means of functional ANOVA, we want to show changes ... -
Correlación entre patrones puntuales en tipos de fuentes de captaciones hídricas en Colombia
In the present study, the relationship between two specific processes is spatially analyzed in order to propose a method to identify the correlation between types of water sources throughout Colombia, the data used ... -
Identificación del modelo TAR cuando el proceso de ruido sigue una distribución generalizada del error y verificación empírica de la estimación de los parámetros
This paper considers the identification of autoregressive orders of the TAR (Threshold AutoRegressive) model assuming the other structural parameters known based in the TAR package of Zhang and Nieto (2017), adapting it ... -
Estimación e imputación de datos faltantes mediante métodos de interpolación espacial para precipitación mensual acumulada en el departamento de Antioquia durante el periodo 2014-2018
Missing values are common in the databases we work on a daily basis, knowing what to do with those missing data is essential, sometimes the immediate solution is to remove the records and lose information that can be of ... -
Modelamiento de la severidad para automóviles asegurados en Bélgica por medio de mixturas finitas de distribuciones para el año 1997
In this work an analysis of the severity for insured cars will be carried out, it is intended to observe the relationship with the type of car, insurance coverage, age of the car, age of the person, gender, group insurance, ... -
Estimación de resultados de encuentros de la liga Santander haciendo uso de un modelo lineal generalizado mixto
This article proposes to use a mixed Poisson model in order to estimate the results of spain soccer league and understand the effects of playing against the different teams that make up the league.the covariates are the ... -
Estimación de la tasa de desempleo e ingreso medio del departamento de Cundinamarca utilizando estimación en áreas pequeñas
In this article, we propose using principal components analysis in a context of high-dimensional spaces to obtain orthogonal variables to use as auxiliary information in Fay-Herriot (FH) and spatial Fay-Herriot (FH). The ... -
Modelo logístico con datos funcionales aplicado al churn en un operador móvil
This document is intended to illustrate the use of multivariate functional data with descriptive analysis, hierarchical cluster implementation and the multiple functional logistic model applied to the Churn prediction of ...