Research on the Factors Affecting House Price using Multiple Linear Regression Model

Abstract

This article aims to identify the factors that affect the change in house prices. In this paper, the Multiple Linear Regression Model is used to analyze the factors with 1000 random samples from the USA which was collected from the 2nd of May in 2014 to the 10th of July in 2014. Based on the datasets, 13 variables (bedrooms, bathrooms, sqft_living, sqft_lot, floors, waterfront, view, condition, sqft-above, sqft_basement, yr_built, yr_renovated, street, city) are selected as independent variables, and used in scatter diagram to show the correlation with the dependent variable price. Turns out that 3 variables have not shown a significant correlation with the price (yr_built, yr_renovate, city). To find which of the factors have the most significant impact on the change in house prices, VIF value is used to compare the collinearity of those 10 variables. Finally turns out that only 5 variables (bedrooms, view, condition, waterfront, sqft_lot) show the most significant effect on the price of housing. Overall, the main factors affecting the price of housing in the USA were the number of bedrooms and views next to the house, as well as the conditions needed for buying the house.