If there are too many irrelevant variables in a linear regression model, the estimates for the relevant ones are negatively affected. Therefore, it is important both in theory and practice to be able to find the model which includes only the most important variables. Traditional approaches consider information criteria. The method of LASSO has become more popular in the past decade.
Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. J. Royal. Statist. Soc B., Vol. 58, No. 1, pages 267-288)
Zou, H., & Hastie, T. (2005). Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 67(2), 301-320.
Zou, H. (2006). The adaptive lasso and its oracle properties. Journal of the American statistical association, 101(476), 1418-1429.
Kontakt: Chong Liang