租賃與買賣房價之關係：以台中為例

Abstract

租房價格會隨著各種因素而浮動，例如：租房位置、屋齡、內部面積、內部空間配置等等。本文從內政部「不動產成交案件實際資訊資料供應系統」取得台中107年至109年的房屋出租及買賣資料，原始模型以房齡、房屋面積、房間數、廳數、衛浴數作為解釋變數，並以時間和地區作為虛擬變數，觀察各個解釋變數對租房價格的影響。 為了提升預測準確度，我們在延伸模型納入新的解釋變數：各出租房方圓100公尺內的平均單位面積成交價格，以此衡量房屋所在區位的價值，並以地理資訊系統進行計算及演示成果。 我們發現，原始模型已經能捕捉74.9%的租房價格變異；而考慮區位價值後，延伸模型能多解釋約1%的變異。本研究顯示，房屋買賣交易價格平均若增加1%，則租屋價格平均增加0.21%。最後，採用分位數回歸模型的結果顯示：隨著分位點愈高，屋齡、樓板面積、附近房屋買賣成交價對租金的影響也愈大。

The rental price will fluctuate with various factors, such as: rental location, age of the house, internal area, internal space configuration and so on. This study uses the housing rental and sales data in Taichung from 2018 to 2020 from the Ministry of the Interior's “Real Estate Transaction Case Actual Information Supply System”. In the original model, the age of the house, the area of the house, the number of rooms, the number of halls, and the number of bathrooms were set as explanatory variables, and the time and region were treated as dummy variables to estimate the impact of each explanatory variable on the rental price. In order to improve the prediction accuracy, we include a new explanatory variable in the extended model: the average transaction price of the unit area within 100 meters of each rental house, so as to measure the value of the location where the house is located. We also use GIS to calculate and demonstrate the results. We find that the original model already captures 74.9% of the variance in rental prices; the extended model can explain about 1% more of the variance when location value is considered. This study shows that an average increase of 1% in the price of a house buying and selling transaction will lead to an average increase of 0.21% in the price of a rental house. Finally, the results of the quantile regression model show that the higher the quantile, the greater the impact of house age, floor area, and nearby house transaction prices on rent.