> ML_LITERATURE // COX-1958-REGRESSION-MODELS-BINARY-SEQUENCES_v1.0
The Regression Analysis of Binary Sequences
David R. Cox · Journal of the Royal Statistical Society: Series B (Methodological) (1958)
foundational1958foundationalthirdPartyReproduced
Principal Contribution
Formulated the logistic regression model using the log-odds (logit) transformation for binary outcomes.
Operational Relevance
The global standard baseline for regulated probability modeling in credit scoring, clinical trials, and conversion modeling.
Assumptions
- Linear relationship between predictors and log-odds of the binary outcome
- Observations are mutually independent
Limitations
- Suffers from complete separation in small samples without regularization
- Cannot capture non-linear decision boundaries without manual feature interactions