特徵價格法之參數與半參數電腦輔助大量估價（CAMA）模型之研究──台北地區法拍屋住宅市場之實證分析

A Comparison between the Semi-parametric and Parametric CAMA Modeling of Court Auction Residential Housing Market in the Taipei Metropolitan Area

因為大部分法拍屋價格低於市場價格，導致法拍屋有高報酬率，因而吸引投資者進入法拍屋市場。然而，法拍屋與中古屋房價之差異為何？本文使用台北市2001年至2003年之法拍屋及中古屋資料，分別建立參數與半參數電腦輔助大量估價(CAMA)估價模型，找出影響房價之重要因素，另運用地理資訊系統將空間變數加入模型中，找出影響房價之空間變數。實證結果顯示以參數模型預測法拍屋與中古屋房價之差距較大，而以半參數模型預測法拍屋與中古屋房價之差距較小。另於模型中加入空間變數結果顯示，以參數模型與半參數模型預測法拍屋與中古屋房價之差距亦較小，顯示半參數模型較參數模型佳，且於模型中加入空間變數能更準確預測房價。

Court auction real estate in the domestic residential housing market has high rate of return, due to the fact that the auction price is always much lower than the market price. However, what is the price difference between the search and auction markets? What are the factors that affect the final bid price of court auction residences? This study will use the comparison between semi-parametric and parametric modelling to find out the relationship between the attributes of court auction residences in the Taipei Metropolitan Area from 2001 to 2003. In addition, this study also adopts the GIS system to find out which spatial factors affect the level of the final bid price on court auction residences. The empirical results show that the parametric approach might bring bigger price gap between the search market and the auction market, but that the use of semi-parametric modelling for measurement and prediction might narrow the gap. Similar results are obtained by adding spatial factors, in which case both semi-parametric and parametric modelling might reduce the gap. Overall, the semi-parametric modelling with or without spatial factors is better than parametric modelling.