都市綠地具有遊憩、防災、調節氣溫等多重效益。雖然不易對這些個別效益的價值進行評估，但都市級地作爲一種公共財，其整體效益可資本化在周圍不動產的價格上。因此，透過辨認不動產內外部及週邊環境如都市綠地等的特徵做爲解釋變數，爲不動產交易價格建立迴歸模型，可得到不動產與都市綠地等相關特徵所隱含的特徵價格。本研究主要目的即在運用地理資訊系統(GIS)調查方法於特徵價格法來衡量與評估都市綠地美質之效益及影響。本文以羅東鎮都市計畫地區作爲實證地區，將其中的公園與學校視爲都市綠地應用，取2004年的房地產交易筆數，GIS調查方法掌握多樣化的環境資源之基本資料，建立房屋的內、外部及環境特徵資料庫，以利特徵價格法之統計、迴歸及環境資源效益特性分析，研究過程分別建立不動產特徵價格一個一般模型與兩個交互作用模型，由三個模型的F檢定及t檢定可以得知，都市綠地特徵的交互頂變數有助於提高對房價應變數的解釋能力。傳統上大部份運用特徵價格法的文獻多以個別獨立變數估計房價，但卻忽略了交互頂變數改變係數的影響。本研究發現房屋內部特徵中土地使用分區型態與樓板面積的交互頂的設定，可以貼近日常生活對單價及總價關係解釋的經驗，反應消費之偏好，房屋外部特徵中客運總站中心之便利與交通可及性對房屋帶來正面之效應，綠地環境特徵中離最近公園距離與500公尺內綠地總和的交互頂，有助於提高對房價應變數的解釋能力，顯示羅東鎮當前已開闢之都市公園綠地對鄰里環境美質與房價所帶來的正面效益，適當地運用這些不動產與都市綠地的特徵價格模型，可用來估計都市綠地資本化在周圍不動產的價格與整體效益價值。

An urban green area can bring multiple benefits of, such as, recreation, hazard prevention, climate adjustment etc. It is difficult to measure the value for each benefit. However, as a kind of public goods, its benefits will capitalize into the prices of the surrounding real properties. By way of identifying the characteristics of the urban green area, and incorporating them with other characteristics of a real property as explanatory variables, we can build a hedonic price model for real properties and find out the hedonic prices of urban green area characteristics. The purposes of this study are to measure and evaluate the impact and benefit of the urban green area amenity by hedonic price method with geography information system (GIS) survey research. We choose the urban planning district in Lou-dong as the study subject in which the urban green area is defined to be parks and schools. The sample is the real property transactions in 2004. By building one general model and two interaction models, and performing F-test and t-test with the data from the three models, it is proved that the interaction variables of the urban green area characteristics are useful for explaining. Traditionally, most of the literatures use hedonic price method to estimate the housing price with several independent variables. But they neglect the varying parameter effects of the interaction variables. The hedonic price model is estimated by using the housing transaction data and the total green area within a 500 meter radius surrounding the house and the total green area within the urban planning district. The estimated hedonic regression model is then used to estimate the benefits of the green area and to simulate the benefits of various distributions of the additional green area in Lou-dong. By applying the hedonic prices models for urban green area, as a kind of public goods, the benefits of the urban green area will capitalize into the prices of the surrounding real properties, the values of urban green area can be evaluated.