In literature, there are many diagnostic measures of multicollinearity, such as VIF and Tolerance, eigenvalues & eigenvectors, R-square, determinant, pairwise correlation, Confidence Intervals and Standard Errors, Theil's Coefficients among some others. I am interested if there is some other diagnostic tests for detection of multicollinearity with reference.
Can be specifically interesting to your research:
Bagheri, A., Habshah, M., & Imon, R. H. M. R. (2012). A novel collinearity-influential observation diagnostic measure based on a group deletion approach. Communications in Statistics-Simulation and Computation, 41(8), 1379-1396.
Bagheri, A., & Midi, H. (2012). On the Performance of the Measure for Diagnosing Multiple High Leverage Collinearity-Reducing Observations. Mathematical Problems in Engineering, 2012.
Curto, J. D., & Pinto, J. C. (2011). The corrected VIF (CVIF). Journal of Applied Statistics, 38(7), 1499-1507.
Victoria-Feser, M. P. (2011). Robust VIF Regression DJ Dupuis.
Most diagnostics for multicollinearity do not provide a way for any statistical validity. A statistical test of multicollinearity is available from Haitovsky. His approach standardizes the non-constant terms in the X'X-matrix to bound the determinant from 0 to 1. The Haitovsky test statistic is calculated in part from the bounded matrix and is distributed chi-square.
Multicollinearity in Regression Analysis: Comment
Yoel Haitovsky
The Review of Economics and Statistics
Vol. 51, No. 4 (Nov., 1969) , pp. 486-489
I recommend you article:
Friendlya M., Kwana E.
Where's Waldo? Visualizing Collinearity Diagnostics, The American Statistician,
Volume 63, Issue 1, 2009,
http://www.datavis.ca/papers/viscollin-web.pdf
you can use this URL to explain that clearly.
http://www.m-hikari.com/ijcms-2011/9-12-2011/rashwanIJCMS9-12-2011.pdf
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics