Application of Fuzzy Theory to Real Estate Markets - A Case Study in Bade City, Taoyuan County

• Main Authors: Chen Huei-Ru, 陳慧茹
• Other Authors: 蔡明達
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/4wh7sq

碩士 === 國防大學理工學院 === 空間科學碩士班 === 102 === As credit crisis of real estate emerged in the USA in August 2007, price estimation of properties in every country around the world would be leading to more reasonable and cautious conditions. However prices in real estate markets have been frequently disturbed by many factors, coupled with the fact that transactions are not transparently registered, and made the pricing hard to quantitatively analyze. Fuzzy theory and its related methodology applied in this research are the main tools for price estimation of properties in order to create reasonable, credible and fast ways to meet the demands from current real estate markets. After establishment of the evaluation model we compare the prices with those of registered transactions and analyze their trends. It is anticipated that fuzzy-based theory can be utilized to the field of property evaluation and provide novel thoughts to consumers for reasonable purchases. In this study we developed a model for price evaluation of properties using fuzzy-based theory according to various factors. After verification we found that the actual transaction prices are higher than those from the model. It is indicated that the main reasons caused high prices in Bade city are psychological influences of nervous property buyers who are willing to pay much more than the average amounts of deals. By using the model we developed it is possible that we can efficiently and systematically derive price ranges for regional areas using Fuzzy factors and spatial analysis by 3,268 data sets of registered transactions. It is indicated that the highest price difference derived from average price of registered transactions and Fuzzy-evaluated price is 1.76 (10 thousand dollars / per unit) and lowest one 0.19. The root mean square error of registered transactions is ±1.05.