Technical Note - An Expectation-Maximization Algorithm to Estimate the Parameters of the Markov Chain Choice Model.
A. Serdar SimsekHuseyin TopalogluPublished in: Oper. Res. (2018)
Keyphrases
- expectation maximization
- em algorithm
- markov chain
- probabilistic model
- parameter estimation
- markov model
- maximum likelihood estimation
- gibbs sampling
- bayesian framework
- transition probabilities
- monte carlo simulation
- mathematical model
- monte carlo
- maximum likelihood
- objective function
- mixture model
- transition matrix
- gaussian distribution
- likelihood function
- markov chain monte carlo
- k means
- learning algorithm
- monte carlo method
- dynamic programming
- estimate the model parameters
- generative model
- prior information
- parameter space
- gibbs sampler
- posterior density
- markov models
- stationary distribution
- gaussian model
- algo rithm
- finite state
- random fields
- steady state
- least squares
- state space
- hidden markov models
- search space
- optimal solution
- reinforcement learning
- bayesian networks