The Doubly Generalized Weibull Power Series Frailty Distribution

This paper introduces the doubly generalized Weibull power series frailty (DGWPSF) model, an extension of the Weibull–k-truncated power series family incorporating a gamma-frailty term. The model enhances flexibility for lifetime data by accommodating unobserved heterogeneity and latent risk factors in survival and reliability studies. We derive its fundamental properties, including the probability density, distribution, survival, and hazard functions, and highlight notable special cases, including the binomial, Poisson, geometric, and logarithmic models. To address the challenges of parameter estimation, we develop the expectation-maximization algorithm and the Bayesian inference procedures. The DGWPSF framework offers a flexible structure for lifetime data analysis, capturing diverse frailty patterns while improving model interpretability and robustness.