Adaptive Mixture of Student-t Distributions

Provides functions to perform the fitting of an adaptive mixture of Student-t distributions to a target density through its kernel function as described in Ardia et al. (2009) . The mixture approximation can then be used as the importance density in importance sampling or as the candidate density in the Metropolis-Hastings algorithm to obtain quantities of interest for the target density itself.

Adaptive mixture of Student-t distributions

DOI CRAN Downloads Downloads

AdMit (Ardia et al., 2009a) is an R package which provides flexible functions to approximate a certain target distribution and to efficiently generate a sample of random draws from it, given only a kernel of the target density function. The core algorithm fits an adaptive mixture of Student-t distributions to the density of interest, and then, importance sampling or the independence chain Metropolis-Hastings algorithm is used to obtain quantities of interest for the target density, using the fitted mixture as the importance or candidate density. The estimation procedure is fully automatic and thus avoids the time-consuming and difficult task of tuning a sampling algorithm. Full description of the algorithm and numerous applications are available in Ardia et al. (2009a) and Ardia et al. (2009b).

Please cite AdMit in publications:

Ardia, D., Hoogerheide, L., van Dijk, H.K. (2009a).
Adaptive mixture of Student-t distributions as a flexible candidate distribution for efficient simulation: The R package AdMit.
Journal of Statistical Software 29(3), pp.1-32.

Ardia, D., Hoogerheide, L., van Dijk, H.K. (2009b).
AdMit: Adaptive mixture of Student-t distributions.
R Journal 1(1), pp.25-30.


Changes in Version 2.1.3 o fix doc

Changes in Version 2.1.2 o fix vignette

Changes in Version 2.1.1 o Info added in DESCRIPTION

Changes in Version 2.1.0 o Preparation for new release on CRAN

Changes in Version 2.0.3 o Update information

Changes in Version 2.0.2 o DESCRIPTION file adapted to new standard o Vignette modified

Changes in Version 2.0.1 o Compilation for new version on CRAN o New email address o Package's structure modified o Vignette's PDF compressed o Speedup of examples o New version number format

Changes in Version 1-01.06 o Speedup of computations for the examples

Changes in Version 1-01.05 o The argument names of .C etc were changed in 2.13.0 to start with .NAME (to avoid confusion with names= in ...) for back-compatibility unnamed first argument

Changes in Version 1-01.04 o dMit and rMit modified to account of several degrees of freedom; this allows MitISEM to use these functions

Changes in Version 1-01.03

o documentation file modified o CITATION file modified

Changes in Version 1-01.02

o CITATION file modified

Changes in Version 1-01.01

o change in AdMit.R to deal with convergence problems for simple cases.

o documentation and codes modified according to JSS style.

o creation of /doc folder with AdMitJSS.txt and AdMitRnews.txt files.

o package vignette added (in /doc).

o CITATION file simplified.

o JSS paper accepted for publication.

o Rnews paper accepted for publication.

Changes in Version 1-00.04

o file AdMitJSS.txt containing the code of the JSS paper is part of the package.

o documentation of AdMit enhanced (explanations for KERNEL construction).

o add reference for Gelman-Meng.

o modify references in CITATION.

o new 'demo' code.

o 'coda' package is now Suggests.

Changes in Version 1-00.03

o change the version in the function 'onLoad'.

Changes in Version 1-00.02

o first version released.

Changes in Version 1-00.01

o previous versions were 'under construction' versions.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


2.1.8 by David Ardia, 5 months ago

Report a bug at

Browse source code at

Authors: David Ardia [aut, cre, cph] , Lennart Hoogerheide [ctb] , Herman van Dijk [ctb]

Documentation:   PDF Manual  

Task views: Bayesian Inference, Cluster Analysis & Finite Mixture Models, Probability Distributions

GPL (>= 2) license

Depends on mvtnorm

Suggests coda

See at CRAN