AYLIK KONUT SATIŞLARININ MODELLENMESİ VE ANTALYA ÖRNEĞİ
MODELLING MONTHLY HOUSING SALES AND EXAMPLE OF ANTALYA

Summary/Abstract: Forecasting is one of the most important activities that must be undertaken in order to take the strategies and measures that the businesses and the individuals will implement in the future. This process is determining the future demand for a service or product in the most accurate and errorfree manner, and in today's competitive environment it has vital importance for the sustainability of the products or the services of the enterprises they provide. In addition to statistical estimation methods, artificial intelligence techniques are effectively used for demand forecasting. In this study, housing demand in Antalya province were estimated using multi-variable linear regression analysis with the EViews program. Estimations were also made using feed forward backpropagation artificial neural networks for the same data set using the Matlab program, and the validity of the results and network performance were compared against the results obtained with the regression model. According to the performance comparison, regression analysis has 9% mean error whereas artificial neural networks shows 1%. Therefore, for the regional housing sale prediction model artificial neural networks show better results.

• Details
• Contents