Application of statistical modeling techniques for PM10 levels forecast in Bogotá

The air quality in Bogotá, Colombia, especially its PM10 level, has become of increasing concern to local authorities, because of its relation to health risks. A forecast system for PM10 levels is beneficial for the preventive policies of environmental agents. The present paper proposes different forecasting models of particulate matter with three data mining techniques. A set of data from 10 stations including PM10 and environmental values was constructed. Following the analysis of the data, three selection methods for the input variables were implemented: select variables with the assistance of an expert group, and using two automatic selection methods. Having three set of potential variables to use as input, three different forecasting methods were implemented: logistic regression, classification trees and random forest. Finally, the validity of the prediction and a comparative of results is made,to conclude about the best forecast model implemented for Bogotá.

• Autores: Mejía Martínez, Nicolás; Montes Martín, Laura Melissa
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2018
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/60898
• Acceso en línea: http://hdl.handle.net/1992/60898
• Palabra clave: Calidad del aire; Material particulado; Técnicas de predicción
• Rights: openAccess
• License: https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf