Fixed effects maximum likelihood estimation of a flexibly parametric proportional hazard model with an application to job exits

► We extend the fixed effects MLE to a flexibly parametric proportional hazard model. ► FEML performs well with interval-constant baseline hazard and covariates. ► We estimate a job duration model with data for young men. ► We find that FEML eliminates downward biased estimates for endogenous covariates.