Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/111722 
Erscheinungsjahr: 
2015
Schriftenreihe/Nr.: 
Tinbergen Institute Discussion Paper No. 15-042/III
Verlag: 
Tinbergen Institute, Amsterdam and Rotterdam
Zusammenfassung: 
This paper presents the R-package <B>MitISEM</B> (mixture of <I>t</I> by importance sampling weighted expectation maximization) which provides an automatic and flexible two-stage method to approximate a non-elliptical target density kernel -- typically a posterior density kernel -- using an adaptive mixture of Student-<I>t</I> densities as approximating density. In the first stage a mixture of Student-<I>t</I> densities is fitted to the target using an expectation maximization (EM) algorithm where each step of the optimization procedure is weighted using importance sampling. In the second stage this mixture density is a candidate density for efficient and robust application of importance sampling or the Metropolis-Hastings (MH) method to estimate properties of the target distribution. The package enables Bayesian inference and prediction on model parameters and probabilities, in particular, for models where densities have multi-modal or other non-elliptical shapes like curved ridges. These shapes occur in research topics in several scientific fields. For instance, analysis of DNA data in bio-informatics, obtaining loans in the banking sector by heterogeneous groups in financial economics and analysis of education's effect on earned income in labor economics. The package <B>MitISEM</B> provides also an extended algorithm, 'sequential MitISEM', which substantially decreases computation time when the target density has to be approximated for increasing data samples.
Schlagwörter: 
finite mixtures
Student-t densities
importance sampling
MCMC
Metropolis-Hastings algorithm
expectation maximization
Bayesian inference
R-software
JEL: 
C01
C11
C87
Dokumentart: 
Working Paper
Erscheint in der Sammlung:

Datei(en):
Datei
Größe
3.27 MB





Publikationen in EconStor sind urheberrechtlich geschützt.