Forecast Of House Prices Based On Partial Linear Model

In recent years,China's housing prices have skyrocketed,and the housing price issue has become a major concern for the people.The excessive growth of house prices will not only affect people's quality of life,but also affect the steady development of our economy.At present,there are various methods for forecasting house prices,which can be summarized as two types.One is based on the factors that influence the price of houses,and the other is the time series that considers house prices.However,these two methods only include the parametric features of the house price in the prediction,and do not consider the non-parametric features of house prices.Considering that partial linear model contains both parametric and non-parametric parts,this paper attempts to use partial linear model to predict house prices.First of all,this paper introduces relevant theoretical knowledge,mainly including the influencing factors of house prices,the classification of regression models,and principal component analysis.Then the commonly used estimation methods and test methods for partial linear models are given.Finally,this paper selects eleven measurable major influencing factors that affect the average sales price of commercial housing in the country from 2000 to 2015,and uses the principal component analysis to reduce the dimension of the selected eleven variables.The partial linear model and the linear regression model are created based on the reduced dimension data.According to the relevant data in 2016,the average sales price of commercial housing in 2016 is predicted.By comparing the fitting results and prediction results of the partial linear model and the linear regression model,it is found that the partial linear model are better than the linear regression model.