Statistical Inference for Markov Renewal Processes.
Abstract
A Markov Renewal Process is one which records at each time t the number of times a system visits each of a finite number (m) of states up to time t. The system moves from state to state according to a Markov chain, and the time required for each move (sojourn time) is a random variable whose distribution function may depend on the two states between which the move is made. In this paper the author develops a test for the goodness of fit of a hypothetical transition probability matrix for a Markov Renewal Process. The author illustrates this procedure numerically by applying it to a realization of a two-state Markov Renewal Process artificially generated on a computer. In addition, the author considers some Bayesian analysis for Markov Renewal Processes by assuming a matrix beta prior distribution for the transition probability matrix. The report also discusses a special case of this topic and gives an illustration for a two-state Markov Renewal Process. In the final chapter a summary of results is given and some possible future research proglems are indicated. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 15, 1971
- Accession Number
- AD0725565
Entities
People
- Dwight B. Brock
Organizations
- Southern Methodist University