Forecasting mountly housing sales to foreigners with type 1 fuzzy regression functions approach based on ridge regression

n recent years, contrary to classical time series analysis methods, artificial neural networks, fuzzy inference systems and many hybrid methods obtained by using such methods are frequently used in forecasting problems. When the literature on the time series forecasting problem is examined, it is known that fuzzy inference systems methods produce very effective results in the forecasting problems. The fact that many fuzzy inference systems depend on the rule base is one of the most important problems of the application of these methods. Type 1 fuzzy regression functions approach, which does not depend on the rule base and has a simpler structure than many other fuzzy inference systems, is frequently used to solve many problems, including the forecasting problem. Although the Type 1 fuzzy regression functions approach has superior forecasting performance, it is known that this method has the multicollinearity problem in the application process of this method. Type 1 fuzzy regressions approach based on ridge regression proposed to eliminate this problem is both eliminate the multicollinearity problem of Type 1 fuzzy regression functions approach and produces better forecasting results than Type 1 fuzzy regression functions approach. In this thesis, the forecasting of house sales to foreigners was realized for the first time with type 1 fuzzy regression functions approach based on ridge regression, which is a fuzzy inference system method. In order to evaluate the performance of the type 1 fuzzy regression functions approach based on ridge regression, which is used for forecasting of house sales to foreigners, the results obtained from many methods proposed in the literature for forecasting problems were compared. As a result of the evaluations, it was concluded that the forecasting results obtained with the type 1 fuzzy regression functions approach based on ridge regression are better than some other methods in the literature.