整合空間統計技術之土地大量估價方法之研究

THE STUDY OF INTEGRATING SPACE STATISTICS TECHNIQUE INTO LAND MASS ASSESSMENT APPROACH

政府機關向來以土地大量估價推估各宗土地之價格，以作為政府課稅之用。本研究由相關文獻回顧，提出「估價區劃設」與「土地價格之形成」等兩大論點，從「空間」的觀點來探討土地大量估價方法的改進，並將空間統計技術與地理資訊系統整合於土地大量估價之中。基於資料取得及相關研究背景考量，茲選擇「台南市東區、中區與西區」等三行政區作為實證分析區域，以進行本研究之實證。由實證研究，本研究將台南市東區、中區與西區劃設六大估價區，分別就各分區建構地價空間擴張預測模型，推估各估價區內之土地價格。最後，就各估價區之推估結果進行預測分析。整體而言，以「整合空間統計技術之土地大量估價方法」所推估的土地價格，同時兼顧統計檢核與估價品質，可有效率且公平地輔助估價者適當地掌握與空間特質相關的現象與特性，推估符合實際空間變化之土地價格。

The government usually adopts the method of land mass assessment to estimate land prices for its taxation. Based on the literatures, we proposed two new perspectives, 'evaluation area's delimitation' and 'land value's forming'. The improvement of land mass assessment approach was discussed from the point of 'space', and was integrated with spatial statistics technique and GIS. Due to the data accessibility and study background, every parcel land value data of Tainan's east district, middle district and west district were adopted. The results present experiment area, which can be divided into six evaluation areas and established land value spatial expansion prediction model to estimate land price in every evaluation areas. Last, we used the predicted results to make predicted analysis. The results show that the approach has both statistics test and assessment quality when integrating space statistics technique with land mass assessment approach to estimate. The estimation of the land prices, which are better matched the real space can be achieved.