Home Get BMS Documentation Tutorials BMA Resources BMS Blog

BMS Blog

The BMS blog presents comments on updates, tutorials, and descriptions of extending BMS (e.g. to custom priors or panels).
See the table of contents below...


News

BMS 0.3.3 released, Fri, 22 Nov 2013 12:39:23 CET
This is a bugfix update. No visible changes with respect to BMS version 0.3.1.Filed under: News

BMS 0.3.1 released, Fri, 25 Oct 2013 22:26:39 CET
Version 0.3.1 of the BMS package for Bayesian Model Averaging has been released on 5 September 2012. This is a maintenance release for compliance with recent CRAN guidelines. The BMS package therefore is again available on CRAN in addition to bms.zeugner.eu. Changes with respect to Version 0.3.0 Starting with R 2.14, the guidelines for R […]

BMS 0.3.0 Released, Thu, 05 May 2011 22:27:03 CET
Version of 0.3.0 of the Bayesian Model Averaging package BMS has been released. Apart from numerous bugfixes, BMS 0.3.0 includes two main additions: The ability in bms to keep certain fixed regressors to be included in all sampled models The option to calculate predictive densities with function pred.density. Moreover, the interrnal structure has been redesigned […]


Advanced Tutorials

Computing the degree of dependency (jointness) among explanatory variables using BMS, Mon, 23 Jul 2012 22:14:38 CET
Capturing the dependence between explanatory variables in the posterior distribution while implementing a Bayesian analysis is crucial. Taking such a dependence into account reveals the sensitivity of posterior distributions of parameters to dependency across regressors. For instance, if two specific variables are complementary over the model space, we expect relatively higher weights for the models […]

BMS and the Fixed Effects Estimator – A Tutorial, Thu, 26 May 2011 22:36:34 CET
This tutorial illustrates how to use Bayesian Model Averaging (BMA) with panel data using the R package BMS. Contents Introduction Fixed Effects Estimation by Demeaning the Data Fixed Effects Estimation with Dummy Variables Bibliography Downloads A pdf version of this tutorial is available here. Introduction Methods for estimating econometric models with panel data have been […]


Beginner Tutorials

Reproducing the BMA exercise of Fernández, Ley and Steel (2001) in R, Mon, 07 Feb 2011 07:24:39 CET
In their paper Model uncertainty in cross-country growth regressions (Journal of Applied Econometrics 2001), Fernández, Ley and Steel (FLS) demonstrate the use of Bayesian Model Averaging (BMA) for a cross-section economic growth data set. This tutorial shows how to reproduce the results with the R-package BMS. A video tutorial (09:56) summarizes the essential part of […]

Replicating the Body Fat Example from “Bayesian Model Averaging: A Tutorial” (1999) with BMS in R, Mon, 07 Feb 2011 07:22:22 CET
In their paper Bayesian Model Averaging: A Tutorial (Statistical Science 14(4), 1999, pp. 382-401), Hoeting, Madigan, Raftery and Volinsky (HMRV) do an exercise in Bayesian Model Averaging (BMA) at pp.394-397 in estimating body fat data from Johnson (1996): “Fitting Percentage of Body Fat to Simple Body Measurements”; Journal of Statistics Education 4(1). This tutorial shows […]


Customize BMS

The bms function explained…, Fri, 05 Nov 2010 22:24:27 CET
This text explains how the bms function works code-wise and is intended for people who prefer to program customized adjustments of the bms package. Note: This post has been significantly revised in Oct 2014, to ensure its interplay with newer BMS releases. bms is the workhorse function to do the sampling part of Bayesian Model […]

Bayesian Model Averaging (BMA) with uncertain Spatial Effects, Wed, 27 Oct 2010 21:46:08 CET
This file illustrates the computer code to use spatial filtering in the context of Bayesian Model Averaging (BMA). For more details and in case you use the code please cite Crespo Cuaresma and Feldkircher (2010). In addition, this tutorial exists as well in PDF form: web_tutorial_spatfilt.pdf Installing spatBMS To get the code started you need […]

Defining Custom Model Priors in BMS, Tue, 11 May 2010 22:10:30 CET
Bayesian Model Averaging (BMA) allows for any kind of model prior distributions. While the R package BMS has built-in support for several types of commonly used priors, there may be the need for constructing a custom model prior in a particular exercise. This tutorial is for users which already dispose of some experience in BMS. […]