Research on the Price Prediction of Commercial Housing in Beijing

Abstract

In order to determine the best appropriate housing price prediction model, this study constructs a linear regression model, a BP neural network model, and a time series model based on data on commercial housing prices and their affecting variables in Beijing during the past 23 years. A set of 12 independent factors that impact the cost of commercial housing is selected first. Then, three models are created, and their fitting effects are compared and analyzed to forecast the average cost of commercial real estate in Shandong Province. It is shown that the multiple linear regression model is more appropriate for long-term prediction and that the time series model predicts data more recently than the BP neural network model; Therefore, it can be used if the prediction period is longer. Although home prices can be predicted theoretically using these three approaches, more work has to be done to fine-tune the applicable models.