Prediction of Housing Price Using Time Series Analysis and Machine Learning Methods
Hae-Jung Chun · Residential Environment Institute Of Korea · 2020
This study compares the ARIMA, VAR and BVAR models of the time series analysis model and the RNN and LSTM models of machine learning methods for predicting housing price. The variables were apartment sales price index, apartment Chonsei price index, corporate bond yield, industrial production index and construction permit area. The temporal range was from January 2001 to December 2018, divided into the whole period and the period before and after the global financial crisis, and the spatial ranges were nationwide, Seoul, Gangnam, and Gangbuk. As a result, it was found that the predictive power of house price using machine learning was better than the time series analysis model. Overall, the estimation based on the machine learning methods showed that the predicted value and the actual value were very similar. Therefore, The government needs to develop a housing market prediction model using artificial intelligence such as machine learning for accurate housing price prediction.