Home  >>  Archives  >>  Volume 16 Number 1  >>  st0419

The Stata Journal
Volume 16 Number 1: pp. 5-22

Subscribe to the Stata Journal

Regressions are commonly misinterpreted

David C. Hoaglin
Independent consultant
Sudbury, MA
Abstract.  Much literature misinterprets results of fitting multivariable models for linear regression, logistic regression, and other generalized linear models, as well as for survival, longitudinal, and hierarchical regressions. For the leading case of multiple regression, regression coefficients can be accurately interpreted via the added-variable plot. However, a common interpretation does not reflect the way regression methods actually work. Additional support for the correct interpretation comes from examining regression coefficients in multivariate normal distributions and from the geometry of least squares. To properly implement multivariable models, one must be cautious when calculating predictions that average over other variables, as in the Stata command margins.
Buy article (PDF): $14.00 View cart
Buy entire issue (PDF): $28.00

View all articles by this author: David C. Hoaglin

View all articles with these keywords: regression models, added-variable plot, multivariate normal distribution, geometry of least squares, margins command

Download citation: BibTeX  RIS

Download citation and abstract: BibTeX  RIS