Use of R² and Its Shortcomings

R. K. Borah¹ , K. K. Singh Meitei² , S. C. Kakaty³

1. Dept. of Statistics, Manipur University, Imphal-795003, India.
2. Dept. of Statistics, Manipur University, Imphal-795003, India.
3. Dept. of Statistics, Dibrugarh University, Dibrugarh-786004, India.

Correspondence should be addressed to: mr_raju06@rediffmail.com.

Section:Research Paper, Product Type: Isroset-Journal
Vol.4 , Issue.6 , pp.12-16, Dec-2017

CrossRef-DOI:   https://doi.org/10.26438/ijsrmss/v4i6.1216

Online published on Dec 31, 2017

Copyright © R. K. Borah, K. K. Singh Meitei, S. C. Kakaty . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
 

View this paper at   Google Scholar | DPI Digital Library

XML View     PDF Download

How to Cite this Paper

609 Views    297 Downloads    168 Downloads

Abstract :
The coefficient of determination R2 is a general measure of usefulness of the regression model. It shows the percentage of the total variation in the response variable which can be explained by the explanatory variable and is considered as the most commonly used measure of goodness of fit for regression models. It is demonstrated by many statisticians and practitioners that expression for the coefficient of determination is generally not equivalent. However it is widely misused. The primary source of the problem is that except for linear models with an intercept term, the several R2 statistics are not equivalent.

Key-Words / Index Term :
Coefficient of determination, Regression model, Regression analysis, Resistant or robust models

References :
[1] Draper, N. R., and Smith. H. (1981), Applied Regression Analysis (2nded.).New York: John Wiley.
[2] Hahn, G. J. (1973), ‟The Coefficient of Determination Exposed,” Chemtech, 3, 609-612.
[3] Hahn, G. J. (1977), “Fitting Regression Models With No Intercept Term.” Journal of Quality Technology, 9, 56-61.
[4] Hawkins, D. M. (1980), ‟ A Note on Fitting a Regression Without an Intercept Term,” The American Statistician, 34, 233.
[5] Healy, M. J. R. (1984), ‟The Use of R2 as a Measure of Goodness of Fit,” Journal of Royal Statistical Society ,Ser. A ,147 , 608-609.
[6] Kvalseth, T. O. (1985). ‟Cautionary Note about R2,”The American Statistician, Vol. 39, No.4, Part 1(Nov., 1985), pp. 279-285.
[7] Montgomery, D. C. and Peck, E. A. (2003), Introduction to Linear Regression Analysis(3rded.). New York :John Wiley.
[8] Marquardt, D. W., and Snee, R. D. (1974), “ Test Statistics for Mixture Models,” Technometrics, 16, 533-537.
[9] Scott, A. and Wild, C. (1991), “Transformations and R2” The American Statistician, Vol. 45, No. 2 (1991), pp. 127-129.
[10] Theil, H. (1971), Principles of Econometrics, New York: John Willy.
[11] Willet, J. B., and Singer, J. D. (1998), “Another Cautionary Note about R2: Its Use in Weighted Least-Squares Regression analysis,” The American Statistician, Vol. 42, No. 3 (1991), pp. 236-238.