An Estimation of Land Price Function Using a Fuzzy Linear Regression Model

Abstract

There have been many studies about estimation of residential bid rent function. Most of those studies commonly use ordinary regression model as a tool of analysis. In ordinary regression models, the deviations of the data set from the regression line can occur because of measurement errors and/or modeling errors.In this paper, a new approach, that is, a fuzzy regression analysis based on fuzzy set theory is introduced for estimating land price function. In the fuzzy regression model, the deviations between the observed values and the estimated values are assumed to result from the fuzziness of a system structure itself. From this point of view, we use this model for the estimation and obtain the fuzzy number of a parameter vector in a linear land price function which fits well with the land price data which are accumulated along major railway lines in Tokyo metropolitan area. This paper is the first step to explain the vague urban phenomena in which human recognition and decision-making are involved.