Prediction of California House Price Based on Multiple Linear Regression

Abstract

In recent years, the price of real estate continues to increase, which is related to the interests of the people and the society and has become a hotspot issue today. Therefore, it is very important to reasonably predict the price of real estate. This paper uses California house price data to solve the problem of how to predict the average annual sales price of California houses through multiple variables. The main distribution of housing prices is obtained through the data, and the influencing factors are analyzed by linear and lasso regression, including the floor area, the number of rooms, the proportion of low-income people, and educational resources in nearby areas. The research on the influencing factors of housing price is very important and basic for solving a series of problems in the current real estate market. Only when we have a full understanding of the micro mechanism of housing market price formation and the influencing factors of housing price, can we effectively and moderately regulate the price of commercial housing.