Modelling the affecting factors of housing price using GMDH-type artificial neural networks in Isfahan city of Iran

Abstract

Purpose The purpose of this paper is to use Group Method of Data Handling (GMDH)-type artificial neural network to model the affecting factors of housing price in Isfahan city housing market. Design/methodology/approach This paper presents an accurate model based on GMDH approach to describing connection between housing price and considered affecting factors in case study of Isfahan city based on trusted data that have been collected from 1995 to 2017 for every six months. The accuracy of the model has been evaluated by mean absolute percentage error (MAPE), root mean square error (RMSE) and mean absolute error (MAE) in this case. Findings Due to the obtained value of MAPE, RMSE and MAE and also their interpretation, accuracy of modelling the factors affecting housing price in Isfahan city housing market using GMDH-type artificial neural network that has been conducted in this paper, is acceptable. Research limitations/implications Due to limitation of reliable data availability about affecting factors, selected period is from 1995 to 2017. Accessing to longer periods of reliable data can improve the accuracy of the model. Originality/value The key point of this research is reaching to a mathematical formula that accurately shows the relationships between housing price in Isfahan city and effective factors. The simplified formula can help users to use it easily for analysing and describing the status of housing market in Isfahan city of Iran.