A Comprehensive Method with Endogenous Land Price for Forecasting Residential Location and Estimating Neighborhood Benefits

本文构建了一个由土地市场均衡决定内源性地价的住宅用地模型。理论上，土地需求函数用随机效用理论建模；土地供给函数假设有地价、人口和面积限制等变量，均衡地价函数是由土地市场均衡条件诱发的。在参数估计和预测方面，本研究只关注户数需求和均衡地价函数（即不考虑土地供求函数）。因为这样既可以避免土地市场出清条件，又可以避免土地供给功能的规范。相反，我们可以只使用家庭数量和土地价格来计算这些数字。本研究亦以岐阜县为个案，检验本模型的适用性。就每个地区目前的户数和土地价格而言，结果的准确性令人满意，但就1980年和1985年之间的差异而言，结果不可靠。最后，利用等效变差的概念，检验了产权价值法在邻域效益估计中的近似程度。

This study constructs a residential land use model with endogenous land pricedetermined by the land market equilibrium. Theoretically the land demand function is modelled by the Random Utility Theory; the land supply function is supposed to have variables of land price, population and area limits: and the equilibrium land price function is induced by the land market equilibrium conditions.As for its parameter estimation and forecasting, this study pays its attentions to only demand in terms of numbers of households and equilibrium land price functions (i.e. excluding land demand and supply function). Because by doing so we can avoid both land market clearance conditions and specification of land supply functions. Instead we can formulate those by using only numbers of households and land prices.This study also examines the applicability of our model through the case study on Gifu Prefecture.The results are satisfactory in terms of accuracy on the present numbers of households and the land prices in each zone, but unreliable on those differences between 1980 and 1985. Finally by using the concept of Equivalent Variation, this study checks the approximation degree of property value approach on neighborhood benefits estimation.