Dear all,
The new version of AdMit (version 1.01-01) is now available from CRAN.
SUMMARY
The package provides functions to perform the fitting of an adaptive
mixture of Student-t distributions to a target density through its
kernel function. 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. We believe that this approach may be applicable in
many fields of research and hope that the R package AdMit will be
fruitful for many researchers like econometricians or applied statisticians.
MODIFICATIONS
o change in AdMit.R to deal with convergence problems for simple cases.
o the documentation file has been improved (thanks to Achim Zeilis for
comments).
o a package vignette has been added.
o a paper describing the package in detail has been published in the
Journal of Statistical Software: http://www.jstatsoft.org/v29/i03.
Abstract:
This paper presents the R package AdMit which provides functions to
approximate and sample from a certain target distribution given only a
kernel of the target density function. The core algorithm consists in
the function AdMit which fits an adaptive mixture of Student-t
distributions to the density of interest via its kernel function. Then,
importance sampling or the independence chain Metropolis-Hastings
algorithm are 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. The
relevance of the package is shown in two examples. The first aims at
illustrating in detail the use of the functions provided by the
package in a bivariate bimodal distribution. The second shows the
relevance of the adaptive mixture procedure through the Bayesian
estimation of a mixture of ARCH model fitted to foreign exchange
log-returns data. The methodology is compared to standard cases of
importance sampling and the Metropolis-Hastings algorithm using a naive
candidate and with the Griddy-Gibbs approach.
o creation of /doc folder with AdMitJSS.txt and AdMitRnews.txt files
(the R codes used for JSS and Rnews papers).
o CITATION file simplified.
o 'coda' package is now Suggests
REFERENCES
Ardia D, Hoogerheide LF, van Dijk HK (2008). AdMit: Adaptive Mixture of
Student-t Distributions in R. R package version 1.01-01.
URL http://CRAN.R-project.org/package=AdMit.
Ardia D, Hoogerheide LF, van Dijk HK (2009). Adaptive Mixture of
Student-t Distributions as a Flexible Candidate Distribution for
Efficient Simulation: The R Package AdMit. Journal of Statistical
Software, 29(3), 1-32.
URL http://www.jstatsoft.org/v29/i03/.
Hoogerheide LF (2006). Essays on Neural Network Sampling Methods and
Instrumental Variables. Ph.D. thesis, Tinbergen Institute, Erasmus
University Rotterdam. Book nr. 379 of the Tinbergen Institute Research
Series.
Hoogerheide LF, Kaashoek JF, van Dijk HK (2007). On the Shape of
Posterior Densities and Credible Sets in Instrumental Variable
Regression Models with Reduced Rank: An Application of Flexible Sampling
Methods using Neural Networks.
Journal of Econometrics, 139(1), 154-180. doi:10.1016/j.jeconom.2006.06.009.
Hoogerheide LF, van Dijk HK (2008a). Bayesian Forecasting of Value at
Risk and Expected Shorfall Using Adaptive Importance Sampling. Technical
Report 2008-092/4, Tinbergen Institute, Erasmus University Rotterdam.
URL http://www.tinbergen.nl/ discussionpapers/08092.pdf.
Hoogerheide LF, van Dijk HK (2008b). Possibly Ill-Behaved Posteriors in
Econometric Models: On the Connection Between Model Structures,
Non-Elliptical Credible Sets and Neural Network Simulation Techniques."
Technical Report 2008-036/4, Tinbergen Institute, Erasmus University
Rotterdam.
URL http://www.tinbergen.nl/discussionpapers/08036.pdf.
Best regards,
David Ardia (package's maintainer)
Lennart F. Hoogerheide
Herman K. van Dijk
[[alternative HTML version deleted]]
_______________________________________________
R-packages mailing list
R-packages at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-packages
