Bandwidth Selection in an EM-Like Algorithm for Nonparametric Multivariate Mixtures

By Tatiana Benaglia, Didier Chauveau, David R. Hunter. 2011

28 citations recorded by the source corpus.

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Identifying Finite Mixtures of Nonparametric Product Distributions and Causal Inference of Confounders (2013, methodology): “use otherwise their iterative algorithm is not guaranteed not to increase from one iteration to another. As stated also by Chauveau et al. (2010), the fact that they do not use an adaptive bandwidth (Benaglia et al., 2011) can be problematic especially when the distributions of the components dier s”
mixtools: An R Package for Analyzing Mixture Models (2009, methodology): “Benaglia et al. (2009a) also discuss speciﬁc cases of model (7) in which some of the f jb k ( ) densities are assumed to be the same except for a location and scale change.”
Non‐parametric estimation of finite mixtures from repeated measurements (2016, background): “As emphasized by Benaglia et al. (2011), in mixture models it is crucial to allow the bandwidth to be component specific, as the component distributions may be very different.”
Semi-Parametric Estimation for Conditional Independence Multivariate Finite Mixture Models (2015, methodology): “[5], but using here the w ij to weight the data.”
Maximum smoothed likelihood for multivariate mixtures (2011, background): “The current article corrects this shortcoming by introducing a smoothed loglikelihood function and formulating an iterative algorithm with a provable monotonicity property that happens to produce results in that resemble those of Benaglia et al. (2009b) in practice.”
Semiparametric mixtures of regressions (2012, methodology): “Though we do not discuss this issue here, recent work by Benaglia et al. (2011) and Chauveau et al. (2010) does so in detail for related algorithms in a non-regression setting.”