Forecasting House Prices in Shanghai Using Common Factor Analysis and Multiple Linear Regression Analysis

Predicting housing prices is critical for understanding economic trends and making decisions in real estate markets. This is accomplished by identifying important influencers and utilizing prediction models. Predicting these prices is essential for understanding broader economic trends and guiding decisions in real estate markets. The model was trained and evaluated using economic and financial data from the CEInet Statistics Database, with key variables representing major financial, demographic, and economic characteristics included. To estimate housing prices, a combination method combining common factor analysis and multiple linear regression was used. The KMO test result of 0.845 and the scores of Bartlett's Test of Sphericity (2=4296.29, p=0.000) confirmed the data's suitability for factor analysis. Factor analysis reduced the dimension of the dataset, revealing two important components responsible for 80.01% of the volatility in home costs. The multiple linear regression model based on these characteristics was tuned to attain good prediction accuracy. The model predicted housing prices in Shanghai with an accuracy of 94.54%.