Inhomogeneous Markov Survival Regression Models
share
We propose new regression models in survival analysis based on homogeneous and inhomogeneous phase-type distributions. The intensity function in this setting plays the role of the hazard function. For unidimensional intensity matrices, we recover the proportional hazard and accelerated failure time models, among others. However, when considering higher dimensions, the proposed methods are only asymptotically equivalent to their classical counterparts and enjoy greater flexibility in the body of the distribution. For their estimation, the latent path representation of semi-Markov models is exploited. Consequently, an adapted EM algorithm is provided and the likelihood is shown to increase at each iteration. We provide several examples of practical significance and outline relevant extensions. The practical feasibility of the models is illustrated on simulated and real-world datasets.
READ FULL TEXT
