The EM Algorithm and Extensions: Second Edition

By Geoffrey J. McLachlan, T. Krishnan. 2008

980 citations recorded by the source corpus.

Subfields: Civil and Structural Engineering

View Paper: https://www.semanticscholar.org/paper/d5c57378bb2a6aabf4d68d390326d55674ed4c99

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Civil and Structural Engineering

1. Daniel Stegmueller
2. Daniel Straub
3. Paul David McNicholas
4. A. Kamariotis
5. Eric Moulines
6. Camila Borelli Zeller
7. Victor Hugo Lachos
8. Jimmy Olsson

Citations in the corpus

• Robust mixture regression modeling based on scale mixtures of skew-normal distributions (2015, methodology): “For the SSL, Monte Carlo integration can be employed, which yields the so-called MC-EM algorithm; see Wei and Tanner (1990), McLachlan and Krishnan (2008) and Zeller et al. (2011).”
• Model-Based Clustering (2016, methodology): “Extensive details on the EM algorithm are given by McLachlan and Krishnan (2008), and a discussion on stopping rules, with some focus on criteria based on Aitken’s acceleration (Aitken 1926), is given by McNicholas (2016, Section 2.2.5).”
• FITTING MIXTURES OF ERLANGS TO CENSORED AND TRUNCATED DATA USING THE EM ALGORITHM (2015, methodology): “…first introduced by Dempster et al. (1977), is an iterative method used to compute maximum likelihood (ML) estimates when the data can be viewed as being incomplete and direct maximization of the incomplete data likelihood is either not desirable or not possible (McLachlan and Krishnan, 2008).”
• Apples and Oranges? The Problem of Equivalence in Comparative Research (2011, methodology): “This model (with identifying restrictions given by equations (8), (10), and (11)) can be estimated by treating the latent variables as missing data which are estimated using the expectation maximization algorithm (McLachlan and Krishnan 2008).”
• Adaptive sequential Monte Carlo by means of mixture of experts (2011, methodology): “Following McLachlan and Krishnan (2008), the learning rate is set to τ` = 1.”
• How Many Countries for Multilevel Modeling? A Comparison of Frequentist and Bayesian Approaches (2013, methodology): “More specifically, likelihood estimation is carried out using a standard EM algorithm (McLachlan and Krishnan 2008), and integration needed in the probit model is done numerically via adaptive ), running 2 chains for 4000 iterations, 12 I also estimated ML models using restricted maximum likelihood.”
• AN OVERVIEW OF COMPOSITE LIKELIHOOD METHODS (2011, background): “See McLachlan and Krishnan (2008) for a book length exposition.”
• A framework for quantifying the value of vibration-based structural health monitoring (2022, positive): “More specifically, we implement an on-line particle filter [52], which performs Gaussian mixture (GM)-based [53] resampling [54] whenever the effective sample size drops below a user-defined threshold (step 8 in the algorithmic summary of Section 5).”
• Semiparametric mixtures of regressions (2012, methodology): “For this modified algorithm, the ascent property of Equation (21) means that many results that apply to true EM algorithms, as in McLachlan and Krishnan (2008), may apply; however, an exploration of the theoretical convergence properties of this algorithm is beyond the scope of the current article.”
• Numerical Maximisation of Likelihood: A Neglected Alternative to EM? (2014, background): “‘Ordinary’ EM is not entirely straightforward for this problem, as a closed-form M step is not possible (McLachlan & Krishnan, 2008, p. 156).”