Advanced

Applications of Geographic Information Systems as an analytical and visualization tool for mass real estate valuation : a case study of Fontibón district, Bogotá, Colombia

Villarreal Pacheco, Rafael D. (2009) In LUMA-GIS Thesis GISM01 20091
Dept of Physical Geography and Ecosystem Science
Abstract (Spanish)
Los avalúos masivos de propiedades están basados en su mayoría en modelos estadísticos globales que tienen en cuenta las características de las propiedades; sin embargo, los modelos utilizados en Bogotá, no tienen en cuenta variables de localización explícitamente. En este estudio se utilizaron Sistemas de Información Geográfica (SIG) como herramienta analítica para desarrollar una serie de variables de localización y
analizar su impacto en el valor de las propiedades; se crearon variables como la proximidad a un aeropuerto, proximidad a centros comerciales y el acceso a vías principales, se generaron mapas de superficie con los valores de las propiedades y se analizó el efecto de las diferentes variables de localización en el precio de... (More)
Los avalúos masivos de propiedades están basados en su mayoría en modelos estadísticos globales que tienen en cuenta las características de las propiedades; sin embargo, los modelos utilizados en Bogotá, no tienen en cuenta variables de localización explícitamente. En este estudio se utilizaron Sistemas de Información Geográfica (SIG) como herramienta analítica para desarrollar una serie de variables de localización y
analizar su impacto en el valor de las propiedades; se crearon variables como la proximidad a un aeropuerto, proximidad a centros comerciales y el acceso a vías principales, se generaron mapas de superficie con los valores de las propiedades y se analizó el efecto de las diferentes variables de localización en el precio de las propiedades. Cuatro diferentes enfoques se utilizaron con el fin de analizar el efecto de
las variables de localización: coeficiente de Pearson, regresión lineal múltiple, step-wise y se utilizaron modelos locales de estadística (Regresión Geográfica Ponderada o Geographically Weighted Regression GWR) para analizar los comportamientos no estacionarios de las variables.
El SIG como herramienta de visualización demostró ser útil en la construcción de mapas de valor, siendo además fundamental para mostrar e interpretar los resultados de los modelos estadísticos locales. Se encontró que en el área de estudio existe una considerable variación de los parámetros, identificando grupos de tendencias, esto hace que el uso de modelos locales de estadística sea ideal. Los modelos locales mostraron que los valores estadísticos en los grupos de predios más costosos parecen estar bien explicadas por las variables de localización, pero en contraste, para los grupos de predios menos costosos, las variables de localización tienen menos capacidad para explicar la
varianza en los valores de los predios. (Less)
Abstract
The majority of approaches for mass appraisal are based on global statistical models that take into account property characteristics, and, in developing countries, location variables are not often explicitly considered. The present study uses Geographic Information Systems (GIS) as an analytical tool to develop a number of location variables and to
analyze their impact in the value of properties. Variables like proximity to an airport, proximity to leisure facilities, proximity to shopping centres, and access to main roads, among others, are created, and property values estimated. Surface value maps are created and the effect of the different location variables in the price is hypothesized. Four
different statistical approaches are used... (More)
The majority of approaches for mass appraisal are based on global statistical models that take into account property characteristics, and, in developing countries, location variables are not often explicitly considered. The present study uses Geographic Information Systems (GIS) as an analytical tool to develop a number of location variables and to
analyze their impact in the value of properties. Variables like proximity to an airport, proximity to leisure facilities, proximity to shopping centres, and access to main roads, among others, are created, and property values estimated. Surface value maps are created and the effect of the different location variables in the price is hypothesized. Four
different statistical approaches are used in order to analyze the effect of proximity attributes: bivariate Pearson coefficient, multiple linear regression, step-wise procedure, and local spatial statistics (Geographically Weighted Regression-GWR) are used to analyze spatially non-stationary behaviours.
The visualization capabilities of GIS prove helpful in constructing value maps, and GIS also proves to be fundamental to report and interpret the results of local spatial statistical models. The results reveal that the spatial variation of some of the parameters is significant and that there are clusters of tendencies. Given this strong clustered distribution, the use of local spatial statistics, to analyze variations of the different
statistical parameters across the study area, is ideal. The local statistical results show that prices in expensive clusters seem to be well explained by location factors, unlike prices in less expensive clusters that are poorly explained. The main conclusion from this thesis is that, for the study area, low-priced properties are more difficult to replicate with proximity attributes than high-priced properties. (Less)
Please use this url to cite or link to this publication:
author
Villarreal Pacheco, Rafael D.
supervisor
organization
alternative title
Sistemas de Información Geográfica como instrumento para visualizar y generar variables de localización para la valoración masiva de predios : caso de estudio Localidad de Fontibón en Bogotá, Colombia
course
GISM01 20091
year
type
H2 - Master's Degree (Two Years)
subject
keywords
mass appraisal, real estate valuation, GIS, location variables, local statistics
publication/series
LUMA-GIS Thesis
report number
4
language
English
additional info
Dr. Armando Blanco Cruz, Universidad Externado de Colombia.
id
3558944
date added to LUP
2013-02-28 10:41:28
date last changed
2013-02-28 12:55:08
@misc{3558944,
  abstract     = {The majority of approaches for mass appraisal are based on global statistical models that take into account property characteristics, and, in developing countries, location variables are not often explicitly considered. The present study uses Geographic Information Systems (GIS) as an analytical tool to develop a number of location variables and to
analyze their impact in the value of properties. Variables like proximity to an airport, proximity to leisure facilities, proximity to shopping centres, and access to main roads, among others, are created, and property values estimated. Surface value maps are created and the effect of the different location variables in the price is hypothesized. Four
different statistical approaches are used in order to analyze the effect of proximity attributes: bivariate Pearson coefficient, multiple linear regression, step-wise procedure, and local spatial statistics (Geographically Weighted Regression-GWR) are used to analyze spatially non-stationary behaviours.
The visualization capabilities of GIS prove helpful in constructing value maps, and GIS also proves to be fundamental to report and interpret the results of local spatial statistical models. The results reveal that the spatial variation of some of the parameters is significant and that there are clusters of tendencies. Given this strong clustered distribution, the use of local spatial statistics, to analyze variations of the different
statistical parameters across the study area, is ideal. The local statistical results show that prices in expensive clusters seem to be well explained by location factors, unlike prices in less expensive clusters that are poorly explained. The main conclusion from this thesis is that, for the study area, low-priced properties are more difficult to replicate with proximity attributes than high-priced properties.},
  author       = {Villarreal Pacheco, Rafael D.},
  keyword      = {mass appraisal,real estate valuation,GIS,location variables,local statistics},
  language     = {eng},
  note         = {Student Paper},
  series       = {LUMA-GIS Thesis},
  title        = {Applications of Geographic Information Systems as an analytical and visualization tool for mass real estate valuation : a case study of Fontibón district, Bogotá, Colombia},
  year         = {2009},
}