Abstract
This article proposes two adjusted coefficients of determination for logistic regression for measuring the strength of association between the outcome variable and the predictors as an extension of a similar concept in linear regression. The proposed adjusted coefficients correct the overestimation problem of the unadjusted coefficients and are especially useful when the sample size is small or the number of predictors is large. An R program is provided on the Internet.
Original language | English (US) |
---|---|
Pages (from-to) | 161-165 |
Number of pages | 5 |
Journal | American Statistician |
Volume | 57 |
Issue number | 3 |
DOIs | |
State | Published - Aug 2003 |
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- General Mathematics
- Statistics, Probability and Uncertainty