Return to search

土地特徵組合估價模式之研究

國內土地估價技術,面臨兩項問題:一是對同一估價標的物土地,在同一個時間,由兩個以上的估價人員估價,所得到的估價結果常出現頗大的差距。另一是如何判斷某一項影響地價重要因素對地價的影響程度?國內相關文獻,雖嘗試使用複迴歸分析中的一般迴歸及逐步迴歸進行分析,期能有助於解決這兩項問題,但是,研究結果,並不理想。在一般迴歸方面,所建立的土地特徵方程式,變數之間常出現嚴重的線性重合現象,部分自變數的係數估計值的。負符號與土地估價理論或先驅訊息不相符合,整個估價方程式在解釋上有所困難。在逐步迴歸方面,所建立的土地特徵方程式,固然改善了影響地價重要因素之間的線性重合問題,但是,相對的,亦將許多影響地價重要因素排除於土。地估價方程式之外。此外,價方程式中的截距項出現負值,亦難以合理解釋,凡此種種問題,均尚未提出解決的方法。本文之研究目的,在於探討如何解決國內估價所面臨之兩項問題,在理論方面,提出「土地特徵組合。概念,認為土地係透過「土地特徵組合」地價的形成發生作用,而非透過「土地特徵」直接封地價之形發生用。在實證方面,引用多變量分析中之因子分析及群落分析,先對影響地價重要因素進行因子分析,並以群落分析結果輔助判斷因子的適當個數,然後再以諸項因子與買賣實例地價進行一般迴歸及逐步迴歸,建立一次式土地估價方程式。此外,另測試建立相加相乘混合型態的非一次式土地估價方程式,以克服上述問題。最後, 對兩種估價方程式加以比較,說明研究發現並提出建議。 / There are two problems of land valuation technique in our country, one problem is that there always are different values i for valuating the price of land at the same time by different valuers, and the difference between these values is tool much to be explained. Another problem is that it is very difficult to identify or to calculate the algorism which demonstrate the degree of every characteristic of land affecting land price. Although, the study of this topic have motioned that the less multi-collinearity the explanatory variables included in the model , the better the model, they never research the fact how the land characteristics affect the land price when they choose the alternative hedonic price function forms. Sometimes they find the intercept of the model is negative, or the sign (positive or negative) of the coefficient of explanatory variables is contradicting to the theory, they also never suggest the remedial measures to the problems. The purpose of this paper is to study the land valuation model from the combination of land characteristics. Consequently, We can get some satisfactory results about resolving above problems.

Identiferoai:union.ndltd.org:CHENGCHI/A2002000608
Creators張杏端
Publisher國立政治大學
Source SetsNational Chengchi University Libraries
Language中文
Detected LanguageEnglish
Typetext
RightsCopyright © nccu library on behalf of the copyright holders

Page generated in 0.0016 seconds