應用群集分析在暢貨購物中心附近房地產之研究與應用-以「MITSUI OUTLET PARK台中港」為例

Abstract

本研究以「MITSUI OUTLET PARK台中港」設置營運前及開幕後的時間，利用「內政部實價登錄」網站2013起至2019年的房地產交易資料，篩選以「MITSUI OUTLET PARK台中港」為中心半徑五公里內的住宅交易資料作為實證分析對象。透過特徵價格模型，以普通最小平方法當作估計的迴歸模型來探討「MITSUI OUTLET PARK台中港」開幕前後房地產各種不同特徵對於房地產價格的影響，並且使用特徵價格法中與房地產價格相關之重要特徵進行群集分析。本研究使用自組織映射圖進行資料分群，藉此探討各群的特性及各群在「MITSUI OUTLET PARK台中港」開幕前後所受到的影響，並將台灣其他兩間暢貨購物中心「MITSUI OUTLET PARK林口」和「華泰名品城」附近的住宅交易資料使用相同的方法，使用資料視覺化分析處理進行比較與討論，探究暢貨購物中心對其附近房地產所帶來的衝擊與影響。

This paper presents empirical analyses on the transaction data of residential properties within a 5-km radius of Mitsui Outlet Park Taichung Port from before its construction began to after it was opened, collected from the Property Sale Price Registration website of the Ministry of Interior from 2013 to 2019. On the basis of a hedonic pricing model, an ordinary least squares regression model was used to discuss the effects of various real estate characteristics on the prices before and after Mitsui Outlet Park Taichung Port was opened. Moreover, variables related to real estate prices in the hedonic pricing model were employed for a cluster analysis. The data were clustered using a self-organizing map to explore the characteristics of each cluster and how it has been affected by before and after Mitsui Outlet Park Taichung Port was opened. The same method was used in two other outlet malls in Taiwan: Mitsui Outlet Park Linkou and GLORIA OUTLETS using data visualization techniques to compare. In order to investigate the importance and impact of the outlet mall to its residential properties.