This Title All WIREs
How to cite this WIREs title:
WIREs Comp Stat

Linear regression

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Linear regression plays a fundamental role in statistical modeling. This article provides a step‐by‐step coverage of linear models in the order of model specification, model estimation, statistical inference, variable selection, model diagnosis, and prediction. Computation issues in linear regression and intimately relevant extensions of linear models are also discussed. WIREs Comput Stat 2012, 4:275–294. doi: 10.1002/wics.1198 This article is categorized under: Statistical Models > Linear Models Algorithms and Computational Methods > Least Squares

Geometric illustration of least squares estimator.

[ Normal View | Magnified View ]

Diagnostic plot of r(−i) vs. hii for the 1987 baseball salary data. The size of the bubble corresponds to Cook's distance Di.

[ Normal View | Magnified View ]

Partial residual plots.

[ Normal View | Magnified View ]

Plot of jackknife residuals r(−i) vs. ŷi: (a) the case where all the model assumptions are valid, superimposed with a smooth curve from loess smoothing; and (b) the case with unequal error variances.

[ Normal View | Magnified View ]

Illustration of shrinkage estimators in the two‐dimensional case: (a) ridge (q = 2) and (b) lasso (q = 1).

[ Normal View | Magnified View ]

Illustration of the bias‐variance tradeoff.

[ Normal View | Magnified View ]

Geometric illustration of the F test.

[ Normal View | Magnified View ]

Browse by Topic

Algorithms and Computational Methods > Least Squares
Statistical Models > Linear Models

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts