Open AccessModelling House Price Using Ridge Regression and Lasso Regression
Author(s) -
Seng Jia Xin,
Karniza Khalid
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i4.30.22378
Subject(s) - multicollinearity , lasso (programming language) , regression analysis , statistics , regression , multivariate statistics , econometrics , mean squared error , ridge , variance inflation factor , elastic net regularization , factor regression model , regression diagnostic , bayesian multivariate linear regression , mathematics , proper linear model , geography , computer science , cartography , world wide web
House price prediction is important for the government, finance company, real estate sector and also the house owner. The data of the house price at Ames, Iowa in United State which from the year 2006 to 2010 is used for multivariate analysis. However, multicollinearity is commonly occurred in the multivariate analysis and gives a serious effect to the model. Therefore, in this study investigates the performance of the Ridge regression model and Lasso regression model as both regressions can deal with multicollinearity. Ridge regression model and Lasso regression model are constructed and compared. The root mean square error (RMSE) and adjusted R-squared are used to evaluate the performance of the models. This comparative study found that the Lasso regression model is performing better compared to the Ridge regression model. Based on this analysis, the selected variables includes the aspect of house size, age of house, condition of house and also the location of the house.