Regression
Logistic regression
Predicts the probability of a binary outcome from one or more predictors.
Variables
- Binary dependent variableNominal · Ordinal — 1
- PredictorsContinuous · Ordinal — at least 1
Options
- Modelled eventSecond level · First level
What it reports
Per predictor B, SE, Wald, p; OR with a 95% CI; model χ² and pseudo-R².
statsmodels · Open the accuracy report
When does this apply?
For a design in which a categorical outcome with exactly two levels — “recovered” / “not recovered”, say — is modelled on one or more quantitative predictors. The outcome column must be categorical, because the table names the modelled level in words; which level counts as the event is set in the “Modelled event” option. The output is the model’s χ², degrees of freedom and p, and per predictor B, its standard error, the Wald z, p, and the odds ratio with a 95% confidence interval. A categorical predictor is not accepted by this model — you must code it 0/1 yourself.