Talk:Regression model validation

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Statistics (Rated Start-class, High-importance)
WikiProject icon

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

Start-Class article Start  This article has been rated as Start-Class on the quality scale.
 High  This article has been rated as High-importance on the importance scale.
 

This is a poor article. It has nothing at all about estimating the model's ability to predict out of sample, the most important part of any validation and applicable to any type of model. Residual analysis only provides a test of the model's assumptions. Ideally an independent dataset should be used to test the model's predictions. The mean square error of prediction (MSEP) is the commonly used measure of fit to the new data. The PRESS statistic is an attempt to estimate the MSEP without independent data using the leave-one-out cross-validation technique and is given by most regression software. All good textbooks describe these measures. Galltywenallt (talk) 11:12, 22 September 2010 (UTC)

The opening of this article gives not citations and yet appears to be just a loose plagiarisation of the source at http://www.itl.nist.gov/div898/handbook/pmd/section4/pmd44.htm — Preceding unsigned comment added by 60.234.129.70 (talk) 23:38, 24 August 2011 (UTC)

I think there was a note claiming This article incorporates public domain material from websites or documents of the National Institute of Standards and Technology. I have added a link under references. (Msrasnw (talk) 10:28, 2 September 2011 (UTC))