Construction of real estate models for feasibility analysis using regression and fuzzy rules

The feasibility study for launching a new property development is often carried out using simplified or subjective techniques. Careful market analysis should be an essential part of this process, in order to reduce part of the risk involved. More precise price estimations may be obtained using hedo...

Full description

Saved in:
Bibliographic Details
Main Authors: González, Marco Aurélio Stumpf, Formoso, Carlos Torres
Format: Online
Language:por
Published: ANTAC - Associação Nacional de Tecnologia do Ambiente Construído 2008
Online Access:https://seer.ufrgs.br/ambienteconstruido/article/view/3718
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The feasibility study for launching a new property development is often carried out using simplified or subjective techniques. Careful market analysis should be an essential part of this process, in order to reduce part of the risk involved. More precise price estimations may be obtained using hedonic price models, based on sales market data. However, estimating the market value of real estate through this process has also some drawbacks, such as imprecision of the boundaries between different sub-markets. Fuzzy rules may be used for dealing with this type of imprecision, making it possible to improve the precision of market appraisal models. This paper presents an application of fuzzy rule-based systems, which have been adjusted by genetic algorithms. Two fuzzy systems were developed. The first one was based on the property area, and the second one was based on its location. Both were compared to a traditional hedonic regression model. The results have indicated that fuzzy rules may improve the quality of real state market analysis.