General EM algorithm for fitting non-monotone hazard functions from truncated and censored observations.
Stéphane R. A. BardeYoung Myoung KoHayong ShinPublished in: Oper. Res. Lett. (2022)
Keyphrases
- em algorithm
- expectation maximization
- parameter estimation
- mixture model
- maximum likelihood
- gaussian mixture model
- generative model
- likelihood function
- maximum likelihood estimation
- incomplete data
- hidden variables
- gaussian mixture
- log likelihood
- probability density function
- expectation maximisation
- maximum a posteriori
- wide class
- likelihood maximization
- machine learning
- gibbs sampling
- penalized likelihood
- probabilistic principal component analysis