Classification of Cases and Odds Ratio
A classification matrix displays the frequencies of the predicted and observed classification of cases and percentage of correct predictions based on the logistic regression model.
An observation is predicted as 1 if the predicted probability is greater than 0.5 else it is predicted as 0. The format of the classification matrix is as follows:
| Predicted
0 | Predicted
1 | |
| Observed:
0 | ƒ0,0 | ƒ0,1 |
| Observed:
1 | ƒ1,0 | ƒ1,1 |
The odds ratio of the classification matrix is given by:
The log odds ratio is the natural log of the odds ratio as defined above.
The Percent correct is given as the percentage correct for a given observed category i and predicted category j.
