A Note on the Use of R-squared in Model Selection
The use of R-squared in Model Selection is a common practice in econometrics. The rationale is that the statistic produces a consistent estimator of the true coefficient of determination for the underlying data while taking into consideration the number of variables involved in the model. This pursuit of parsimony comes with a cost: The researcher has no control over the error probabilities of the statistic. Alternative measures of goodness of fit, such as the Schwarz Information Criterion, provide only a marginal improvement to the problem. The F-Test under the Neyman-Pearson testing framework will provide the best alternative for model selection criteria.
|Date of creation:||21 Oct 2007|
|Date of revision:|
|Contact details of provider:|| Postal: P.O. Box 8795, Williamsburg, VA 23187-8795|
Phone: (757) 221-4311
Fax: (757) 221-2390
Web page: http://www.wm.edu/economics/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:cwm:wpaper:62. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Daifeng He)or (Alfredo Pereira)
If references are entirely missing, you can add them using this form.