Housing Price Forecasting Using AI (LSTM) | ||
| تحقیقات مالی | ||
| Volume 25, Issue 4, 2023, Pages 557-576 PDF (1.1 M) | ||
| Document Type: Research Paper | ||
| DOI: 10.22059/frj.2023.349924.1007398 | ||
| Authors | ||
| Hossein Ziyadi1; Erfan Salavati* 2; Mohammad Mahdi Lotfi Heravi3 | ||
| 1MSc., Department of Mathematical Finance, Faculty of Mathematics and Computer Science, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran. | ||
| 2Assistant Prof., Department of Mathematical Finance, Faculty of Mathematics and Computer Science, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran. | ||
| 3Assistant Prof., Department of Economics and Finance Faculty of Management, Science and Technology, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran. | ||
| Abstract | ||
| Objective Forecasting the asset prices in any market is an inseparable and critical part of research on markets. Obtaining a credible prediction of an asset's potential future price provides valuable information for investors. Besides that, being confronted with real estate price shocks and price fluctuations in alternative markets, determining the best time for the investment may come up as a big challenge for any investor. Tracing the trend of housing price changes in Iran shows that the average housing price has had a general trend close to other prices and price indices; but the significant issue in this market is the various growth processes and dynamics, compared to changes in other economic indicators. This dynamic gets more complex when a variety of quantitative and qualitative data from diverse types and from several markets are added to the model. This diversity along with the unstructured, stochastic, and scattered datasets makes the implementation of the model so hard to do. The main goal of this paper is to build an Artificial Intelligence model with the most flexibility in the input part for the data type heterogeneity and the lowest error in the output part as the predicted price. All the implemented models showed a high accuracy on the real data extracted from the housing market. Methods Artificial Intelligence models have an enormous capacity to get a broad range of any kind of data and process and concurrently compile them to build a valid output. This characteristic of AI is very useful in financial models to increase the accuracy of output. Our model is based on the Recurrent Neural Networks. Due to its capability to preserve past information, the LSTM algorithm was implemented as a time series forecasting model. Results In this study, using diverse types of official datasets, such as the Central Bank of Iran, we distinguish the influencing variables in the housing market and then we could predict the average housing price in Tehran. Our findings indicate that the average housing price demonstrates the strongest correlation with gold prices, foreign exchange rates, the Consumer Price Index, and market liquidity levels. Utilizing these indicators, predictions with very high accuracy were obtained. Conclusion Among the four different models of this research, the best prediction belonged to the multivariate stacked-LSTM model, which was empowered by the highly correlated macroeconomic variables. The validation of models was done by the MAPE Indicator. Furthermore, all the results confirmed that the LSTM algorithm was highly effective in utilizing over two decades of actual housing data from Tehran to forecast future prices. | ||
| Keywords | ||
| Housing price; LSTM; Price prediction; Time series forecasting | ||
| References | ||
|
Abbasinejad, H. & Teshkini, A. (2005). Is inflation a monetary phenomenon in Iran? Journal of Economic Research, 39(4), 181-212. (in Persian) Bishop, C.M. (2006). Pattern recognition and Machine Learning. Springer Press, 225-290. Central Bank of Iran (1993-2022). National Economic Indices Periodic Survies. Official Publications of CBI. (in Persian) Chen, X., Wei L. & Xu, J. (2017). House price prediction using LSTM. The Computing Research Repository (CoRR-2017). Eslami Bidgoli, G. & Bajalan, S. (2008). Test of the quantity theory of money in Iran and examination of the effectiveness of price stabilizing policy with GARCH models. Economics Research, 8(29), 205-225. (in Persian) Eyvazloo, R., Eslamibidgoli, S. & Khorsandi, A. (2019). Comparing repeated sales indices (BMN and Case-Shiller) in real estate markets in city of Tehran. Financial Research Journal, 21(3), 348-363. (in Persian) Gholizade, A.A. & Mollavali, T. (2012). The effects of liquidity on housing price fluctuations in oil-producing countries vs. other countries. Quarterly journal of economic research and policies, 20 (63), 83-104. (in Persian) Goldberg, K. P. & Knetter, M. M. (1997). Goods prices and exchange rates: what have we learned? Journal of Economic Literature, 35(3), 1243-1272. Goodfellow, I., Bengio, Y. & Courville, A. (2016). Deep Learning. MIT Press. Hansson, F. & Rostami J. (2019). Time series forecasting of house prices: an evaluation of a support vector machine and a recurrent neural network with LSTM cells. Uppsala University Thesis. May 2019. Heidari, M. & Amiri, H. (2022). Inspecting the Predictive Power of Artificial Intelligence Models in Predicting the Stock Price Trend in Tehran Stock Exchange. Financial Research Journal, 24(4), 602-623. (in Persian) Hosseini, S. S. & Mohtashami, T. (2008). The relations of money growth and inflation in Iran economy; interruption or satiable? Quarterly Journal of Economic Research, 8(3), 21-42. (in Persian) Jahangiri, Kh. & Hoseini Ebrahimabad, S.A. (2022). The Study of Monetary Policy, Exchange Rate and Gold Effects on the Stock Market in Iran Using MS-VAR-EGARCH Model. Financial Research Journal, 19(3), 389-414. (in Persian) Karevan, Z., Johan A.K. & Suykens, J. (2019). Transductive LSTM for time-series prediction: an application to weather forecasting. The official journal of the International Neural Network Society, 125(7567). Katiyar, S. & Borgohain, S. (2021). Image captioning using Deep stacked LSTMs, contextual word embeddings and data augmentation. National Institute of Technology, India (Preprint submitted to ArXiv on Feb 2021). Limsombunchai, V., Gan Ch. & Lee, M. (2004). House price prediction: Hedonic price model vs. ANN. American Journal of Applied Sciences, 1 (3), 193-201. Mousavi, M. & Doroodian, H. (2016). The determinants of housing prices in Tehran. Quarterly Journal of Economical Modeling, 9(31), 103-127. (in Persian) Rabbya, F., Tua, Y., Hossena, I., Lee, I., Maidaa, A. S. & Hei, X. (2021). Stacked LSTM based deep recurrent neural network with Kalman smoothing for blood glucose prediction. The University of Louisiana at Lafayette, Lafayatte & University of Pennsylvania, Philadelphia, USA (Preprint submitted to ArXiv on Jan 2021). Shams, Sh. & Naji Zavareh, M. (2015). Comparison Between the Hybrid Model of Genetic Fuzzy and Self - Organizing Systems and Linear Model to Predict the Price of Gold Coin Futures Contracts. Financial Research Journal, 24(4), 602-623. (in Persian) Tsatsaronis, K. & Zhu. H. (2004). What drives housing price dynamics: cross-country evidence. Bank for International Settlements (BIS) Quarterly Review, 65-78. Statistical Center of Iran (1993-2022). Economic Indices and Households Income and Expenditure Periodic Survies. Official Publications of SCI. (in Persian) Xiao, F. (2020). Time series forecasting with stacked Long Short-Term Memory networks. Presented for Toronto Transit Commission Nov 2020. Yu, L., Jiao, C., Xin, H., Wang, Y. & Wang, K. (2018). Prediction on housing price based on Deep Learning. International Journal of Computer and Information Engineering, 12(2), 90-99. Zhang, H. Li, L. Hui, E. Ch. Li, V. (2016). Comparisons of the relations between housing prices and the macroeconomy in China’s first. second- and third-tier cities. Habitat International. 57, 24-42. Zobeiri, H. & Nademi, Y. (2015). Exchange rate gap effect on unemployment rate in Iran using Markov-switching model. The Quarterly Journal of Planning and Budgeting, 20 (1), 109-136. (in Persian) | ||
|
Statistics Article View: 1,792 PDF Download: 1,105 |
||