iterLap: Approximate probability densities by iterated Laplace Approximations

Share:

The iterLap (iterated Laplace approximation) algorithm approximates a general (possibly non-normalized) probability density on R^p, by repeated Laplace approximations to the difference between current approximation and true density (on log scale). The final approximation is a mixture of multivariate normal distributions and might be used for example as a proposal distribution for importance sampling (eg in Bayesian applications). The algorithm can be seen as a computational generalization of the Laplace approximation suitable for skew or multimodal densities.

Author
Bjoern Bornkamp
Date of publication
2012-05-22 20:52:08
Maintainer
Bjoern Bornkamp <bbnkmp@gmail.com>
License
GPL
Version
1.1-2

View on CRAN

Man pages

GRApprox
Gelman-Rubin mode approximation
ISandIMH
Monte Carlo sampling using the iterated Laplace...
iterLap
Iterated Laplace Approximation
iterLap-internal
iterLap package internal functions
iterLap-package
iterLap package information
resample
Residual resampling

Files in this package

iterLap
iterLap/MD5
iterLap/DESCRIPTION
iterLap/R
iterLap/R/iterLap.R
iterLap/inst
iterLap/inst/CITATION
iterLap/src
iterLap/src/iterLap.c
iterLap/NAMESPACE
iterLap/man
iterLap/man/iterLap-internal.Rd
iterLap/man/resample.Rd
iterLap/man/GRApprox.Rd
iterLap/man/iterLap.Rd
iterLap/man/ISandIMH.Rd
iterLap/man/iterLap-package.Rd