Research on the Influencing Factors and Prediction of Housing Prices Based on Regression Analysis --- Taking California as an Example

Abstract

A house is a necessity for everyone's life. But in today's world, with the continuous reduction of land area, the continuous increase of population, and the continuous maturity of the real estate industry, people have to consider more carefully what factors are the most important in the house. Based on the background of housing prices in California, this paper uses linear regression, random forest and principal component analysis to determine which variables have the greatest impact on housing prices. The reason for using three methods is to obtain more accurate results. According to the results, linear regression shows that income is the most relevant variable to housing prices. Random Forest shows an R square of 75.7%, meaning the predictions fit the data fairly well. Principal component analysis also shows that income is the most important variable. At the same time, house prices will be predicted based on the data obtained, which is about 116598.48588. Some suggestions for those who want to buy a house - which factors of the house are more important. Second, this article will also analyse how to face the rising housing prices.