Shanghai Second-hand Real Estate Price Prediction

Abstract

With the rapid urbanization and continuous growth of the real estate market, housing price prediction has become an important research topic in urban economics. This paper studies second-hand housing prices in Shanghai using data from 2022 to 2024. The dataset includes 16 districts and covers factors such as housing size, location, distance to subway stations and school resources. After removing incomplete data 3126 samples were kept for analysis. Both statistical methods and machine learning models were used, including linear regression and random forest. The results suggest that housing prices vary a lot across regions. Central districts are clearly more expensive while outer areas are relatively cheaper. There is also a slight downward trend over time. Compared with other models, random forest performs better, especially when handling more complex relationships. Factors like location and transport seem to play an important role. However, some limitations remain such as missing macro-level data. Future research could improve this by including more variables and trying other models.