No CrossRef data available.
Article contents
Linear Regression in High Dimension and/or for Correlated Inputs
Published online by Cambridge University Press: 23 January 2015
Abstract
Ordinary least square is the common way to estimate linear regression models. When inputs are correlated or when they are too numerous, regression methods using derived inputs directions or shrinkage methods can be efficient alternatives. Methods using derived inputs directions build new uncorrelated variables as linear combination of the initial inputs, whereas shrinkage methods introduce regularization and variable selection by penalizing the usual least square criterion. Both kinds of methods are presented and illustrated thanks to the R software on an astronomical dataset.
- Type
- Research Article
- Information
- Copyright
- © EAS, EDP Sciences, 2015
References
Hastie, T., Tibshirani, R., & Friedman, F., 2009, The Elements of Statistical Learning, Data Mining, Inference and Prediction, 2nd edition (Springer)Google Scholar
Hoerl, A.E., & Kennard, R., 1970, Ridge Regr.: Biaised Estim. Nonorthog. Probl., Technom., 12, 55
Lawless, J.F., & Wang, P., 1976, Comm. Stat., Theory Meth., 14, 1589
Tibshirani, R., 1996, J. Royal Stat. Soc. Series B, 58, 267
Tibshirani, R., Saunders, M., Rosset, S., Zhu, J., & Knight, K., 2005, J. Royal Stat. Soc. Series B, 67, 91CrossRef
Yengo, L., Jacques, J., & Biernacki, C., 2014, J. Soc. Française Stat. (in press)
Zhou, H., & Hastie, T., 2006, J. Royal Stat. Soc. Series B, 101, 1418