The Modelling of Residential Sales Prices with Kriging Using Different Distance Metrics in Different Correlation Functions

Abstract

The modelling and estimation of sales prices based on economical conditions are important for housing sector especially in developing countries. Analysts are focused on the subject to analyze price movements and estimate the future trend of the sales prices for housing sector. In this study, we tried to generate a robust and efficient model related to the subject. Firstly, we investigated economic variables affecting housing sales prices and then created a kriging model for housing sales prices in Dubai. To determine a better correlation function structure for creating a powerful kriging model, we used Euclidean and Canberra distances for both Exponential and Gaussian correlation functions. Simulation studies were applied to obtain optimum correlation functions. To detect the normality of response values, Focused Information Criteria(FIC) was used. By using cross validation criteria, we selected the best performed correlation function with the best performed distance metric for the kriging model.