Machine Learning-based Models for House Price Prediction in Provincial Administrative Regions of China
These authors contributed equally.
- https://doi.org/10.2991/aebmr.k.220307.035How to use a DOI?
- Machine learning; House price prediction; Regression analysis
House price is an intractable problem with numerous influence factors. In this paper, boosting and traditional algorithms are compared to screen out the optimal model for house price prediction in provincial administrative regions of China. Based on provincial house price data in China from 2000 to 2019, the data is preprocessed by doing statistical analysis, dealing the missing values as well as choosing characteristic features for analysis. Then the data is imported into different models, comparing the prediction effects to pick out the best and then optimizing the hyper-parameters. Using mean squared error (RMSE), root mean absolute percentage error (MAPE), R-square (R2), and explained variance score (EV) as evaluation indicators to appraise the models, the result presents that CatBoost is better than any other models, whose MAPE is 12.5%, R2 is 87.81% and EV is 90.5%. Then sub-sample test is used to examine the robustness, whose result shows that CatBoost is always effective. The empirical findings mainly show that CatBoost is effective in predicting house prices with complex variables and the feature importance graph generated by CatBoost presents that demand and macro environment factors can explain the major fluctuation of house price and that in macro environment factors, macro-economic and education indicators are obviously important than other macro indicators.
- © 2022 The Authors. Published by Atlantis Press International B.V.
- Open Access
- This is an open access article under the CC BY-NC license.
Cite this article
TY - CONF AU - Xiafei Ding AU - Weiya Wang AU - Yiqian Zhang AU - Xiaoyuqian Zhong PY - 2022 DA - 2022/03/26 TI - Machine Learning-based Models for House Price Prediction in Provincial Administrative Regions of China BT - Proceedings of the 2022 7th International Conference on Financial Innovation and Economic Development (ICFIED 2022) PB - Atlantis Press SP - 221 EP - 229 SN - 2352-5428 UR - https://doi.org/10.2991/aebmr.k.220307.035 DO - https://doi.org/10.2991/aebmr.k.220307.035 ID - Ding2022 ER -