Multidimensional house price prediction with SOTA RNNs

• Published in: International journal of strategic property management Vol. 28; no. 6; pp. 411 - 423
• Main Author: Kutuk, Yasin
• Format: Journal Article
• Language: English
• Published: Vilnius Vilnius Gediminas Technical University 01.11.2024
• Subjects: Business forecasting; deep learning; Dwellings; Housing; housing price index prediction; Housing prices; Marketing research; Methods; Neural networks; Prediction theory; Prices and rates; recurrent neural networks; REITs; Residential real estate; Turkey; recurrent neural networks; deep learning; housing price index prediction
• ISSN: 1648-715X; 1648-9179; 1648-9179
• DOI: 10.3846/ijspm.2024.22661

This paper introduces insights into the Turkish real estate market, which can be generalized globally. It primarily aims to find the best forecasting algorithms for the housing price index and compare their prediction performance over three, six, nine, and twelve months ahead by using recurrent neural networks (RNN) with a comparison of out-of-sample predicting power of econometrical models. For these purposes, we employ three RNN architectures in twenty-four settings, revealing that certain RNN architectures are the best predictors in forecasting the Turkish real housing price index. The RNN architectures outperform traditional econometric models; however, the more months forecasted, the lower the prediction power. The lagged values of the price-to-rent ratio, real rents, and the lagged USDTRY values contribute more than the other predictors in forecasting the real housing price index. The outcomes suggest that stocks, real estate investment trusts, and gold are neither complementary nor competing financial instruments since housing is an illiquid asset.