Applied Linear Regression

by Sanford Weisberg

A practical reference and advanced undergraduate and first year graduate text that outlines the methodology for fitting linear models to data. Emphasizing model building, assessing fit and reliability, and drawing conclusions, it uses real data examples and problems to illustrate simple linear regression, multiple regression, weighted least squares, residuals and influence, symptoms and remedies, defining new predictors, collinearity and variable selection, prediction, incomplete data, non-least squares estimation, and more. Includes real data examples and problems and a table of critical values of a test for outliers.

Genres: Mathematics, Reference, Textbooks

324 pages, Hardcover

First published January 1, 1980

Reviews

[Shelby Gonzalez · Rating 3 out of 5]

Read for my stats class on linear regression. Definitely hard to stay awake while reading sometimes

[RA · Rating 4 out of 5]

I read the fourth edition and was happy to learn a number of things that were new to me. Not everything was easy to understand or complete, but I was not disappointed, it is a book worth reading, although as for most other statistics books, one cannot expect to cover everything.

[Shika · Rating 3 out of 5]

Assumes you have linear algebra/strong mathematical background. If you're a biologistic looking to starting in Biostatistics, would recommend looking elsewhere. If you're a statistician looking for applied linear regression basics, then this would be better suited for you.

[Goo · Rating 2 out of 5]

I read some of this as a secondary text when first learning the theory of linear normal models. My primary text lacked certain aspects of actually applying linear models which this text covered in a simple fashion. However, the text is not suited as an introduction to the mathematical theory behind linear models.

[Patrick · Rating 2 out of 5]

I read this as a graduate student in statistics, there are many better books.