Logistic Regression Research: 99% Concordance, Few Significance

Click For Summary
SUMMARY

The discussion centers on the use of logistic regression in SAS, achieving a concordance rate of 99%. However, the analysis reveals that very few variables are statistically significant. This indicates a potential issue with model overfitting or the need for variable selection techniques to improve the model's interpretability and predictive power.

PREREQUISITES
  • Understanding of logistic regression methodology
  • Familiarity with SAS programming and output interpretation
  • Knowledge of statistical significance and p-values
  • Experience with model evaluation metrics, particularly concordance
NEXT STEPS
  • Explore variable selection techniques in SAS, such as stepwise regression
  • Learn about model diagnostics and validation methods for logistic regression
  • Investigate the implications of high concordance with low significance
  • Study advanced logistic regression techniques, including regularization methods
USEFUL FOR

Data scientists, statisticians, and researchers involved in predictive modeling and logistic regression analysis will benefit from this discussion.

mathmathRW
Messages
8
Reaction score
0
I am doing some research and running a SAS program using logistic regression. The concordance is 99%, but hardly any variables are significant. Can anyone help me understand what this means?
 
Physics news on Phys.org
Hey mathmathRW.

Can you please post your output data from SAS so we can take a look at it?
 

Similar threads

Replies
3
Views
3K
  • · Replies 14 ·
Replies
14
Views
3K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 30 ·
2
Replies
30
Views
5K
  • · Replies 23 ·
Replies
23
Views
4K
  • · Replies 21 ·
Replies
21
Views
3K
  • · Replies 64 ·
3
Replies
64
Views
6K
  • · Replies 8 ·
Replies
8
Views
3K
Replies
1
Views
1K