Predicting the Rise in California Home Prices and Factors Affecting

Abstract

California’s real estate market has been in the spotlight for its unique geography and economic advantages, and home prices have been volatile. Therefore, accurate prediction of house price increases is of great importance to home buyers, investors and policy makers. This paper utilizes the California house price dataset, combines macroeconomic indicators and micro property attributes, and analyzes house price prediction through random forest, decision tree and neural network models. The study results show that the Random Forest model performs the best in predicting house prices with an R²value of 0.817210, which is significantly higher than the other models at 0.817210. the influence of R²leads to a slightly poorer neural network performance (CNN). Further through visualization studies and feature importance analysis median income, longitude, latitude, and distance from the sea are the key factors affecting the house price in California. House prices are strongly positively correlated with median income, while distance from the sea is negatively correlated with house prices. These key factors can also directly affect future home buying trends and house price direction. The research in this paper provides a strong reference for homebuyers, investors, and policymakers to help them make more informed decisions in the real estate market and promote a smooth and healthy market development.