Some Applications of Phase-type Distributions in Recurrent Events
نویسندگان
1 Department of Mathematics, Oregon State University, Corvallis, OR, USA.
2 Department of Statistics, Shahid Beheshti University, Tehran, Iran
doi
10.22034/jirss.2024.1999145.1011چکیده
In this paper, the recurrent events that can occur more than once over the follow-up time have been modeled by phase-type distributions. The time till death is assumed to have a phase-type distribution (defined in a Markov chain environment) with interpretable parameters. We use phase-type distribution to calculate the probability of various number of transitions, the conditional expected time to stay in a disease stage and the probability of transition from a stage to another. Using both real and simulated datasets, the model has been calibrated. The confidence intervals for the parameters have been created using bootstrap approaches.