The Proportional Hazards Assumption (PHA) in Cox Models is a fundamental condition stating that the hazard ratio between any two individuals remains constant over time, regardless of their entry into the study or elapsed duration. Formally defined by the requirement $h(t|x_1)/h(t|x_2) = \text{constant}$ where hazards are modeled as an exponential function of covariates multiplied by a baseline hazard independent of those specific variables, this mechanism operates within semi-parametric survival analysis to decouple subject-specific risk factors from the underlying temporal distribution of events. This concept serves as the primary methodological constraint allowing partial likelihood estimation without specifying the baseline hazard curve, distinguishing it fully parametric or non-proportional models in statistical inference domains.
Prerequisite chain context: requires Cox Regression in Survival Analysis.