Determining of variables affecting house prices in Aydın province through fuzzy regression analysis

In Turkey, housing and construction sector, especially after 2002, has become one of the most important sectors of economic growth. In recent years, studies of determining the factors affecting housing demand, estimating house prices and determining the effect of price on demand have frequently appeared in the literature. In empirical researches to determine the housing prices, it is generally observed that classical regression analysis methods are used. In classical regression analysis, the relationship between dependent variable and independent variable or variables is estimated based on observed data. Problems in which classical regression analysis is insufficient due to insufficient number of observations, difficulties encountered while testing the accuracy of assumptions and uncertainty in the relationship between dependent and independent variables can be solved with fuzzy regression analysis. In fuzzy regression analysis, the error resulting from variables or measurement errors, known as the error term in classical regression, which are likely to explain the dependent variables but cannot be included in the model, is distributed to all model coefficients. In this study, it is aimed to determine the characteristics of housing that affect the housing prices, which is the most basic determinant of housing demand, with the data related to the sample of Aydın. In the estimation of the housing price, fuzyy least squares method has been used. The study is powerful than similar studies with this aspect, and it is expected to make a significant contribution to the relevant literature, provincial economy and studies in this field. KEYWORDS: Housing prices, Housing Demand, Fuzzy Logic, Fuzzy Regression