×
Understanding Logistic Regression
What is Logistic Regression?
Logistic regression predicts binary outcomes (disease present/absent) using predictor variables. It outputs probabilities between 0 and 1 using the sigmoid function.
How Coefficients Work
- Positive coefficients: Increase risk as feature value increases
- Negative coefficients: Decrease risk as feature value increases
- Large absolute values: Stronger influence on outcome
- Near-zero values: Minimal impact
Why Interpretability Matters
In healthcare, clinicians need to understand why a model makes predictions to trust and act on them. Feature contributions help identify which risk factors are most important for individual patients.
Formula
linear = m₁×x₁ + m₂×x₂ + ... + bias
probability = 1 / (1 + e-linear)