Link to the last RSS article here:
Installing SPSS 14 - Ed.
Bayesian Packages for R version 2.2.1
By Dr
Rich Herrington, ACS Research and Statistical Support Services
Consultant
R version 2.2.1 was released on December
20, 2005. A description of changes for Microsoft Windows platform
versions can be found
here. The latest versions of R for the Windows platform can be
found here.
We are hosting a local copy of the R 2.2.1 installation file for the
Windows platform here
for download. Additionally, we have made free copies of a
"executable" CD version of R 2.2.1 available over in the UNT bookstore (trade
books).
R 2.2.1 has been installed on this "live" CD with some minor
pre-configurations so that R can run off of the CD. To begin the R
session, browse to the \bin folder on the CD and click the
Rgui.exe file. Additionally, you can set up a shortcut on your
desktop to run R off of the CD. Alternatively, you can copy the
contents of the CD into an R folder on your local hard-drive and create
a shortcut to the C:\R\bin\Rgui.exe file. Once R has been
started, you should see two windows once the initialization of R is
complete. The first is the R Console window:
The second is the
R Commander window, a simple drop down menu
for doing a number of beginner to advanced statistics thru a GUI dialog
system:
To set up your working directory so that you can read and write files
to your local storage media (not the CD - since it is write only).
Go the R console window and select "File - Change dir":
You can set your working directory to a writeable storage media:
Now you should be able to write out and read to this directory.
Bayesian Analysis in R
Bayesian
approaches to inference have become increasingly popular in applied
statistics since the arrival of cheap, fast, computers. The
availability of today's computational power in a desktop PC allows more
complicated, realistic Bayesian models to be estimated thru
simulation
methodologies. Currently, the availability of the
BUGS software
(e.g.
WinBugs &
OpenBugs) and numerous
R
packages dedicated to Bayesian analysis, give the researcher
an arsenal of methods to attack problems from a Bayesian framework.
An introduction to Bayesian thinking and data analysis is beyond the
scope of this current article, but we hope to cover an introduction to
Bayesian analysis in R and WinBugs in future column installments.
Here, we just present a sampling of some of the packages that are
available in R for doing Bayesian analysis:
R2WinBUGS Running WinBUGS from R
bayesm
Bayesian Inference for Marketing/Micro-econometrics
bayesSurv Bayesian
Survival Regression with Flexible Error and Random Effects Distributions
BayesTree Bayesian Methods
for Tree Based Models
baymvb
Bayesian analysis of multivariate binary data|
BMA
Bayesian Model Averaging
boa
Bayesian Output Analysis Program (BOA) for MCMC
deal
Learning Bayesian Networks with Mixed Variables
ebayesthresh Empirical Bayes thresholding
and related methods
eco
R Package for Fitting Bayesian Models of Ecological Inference in 2x2
Tables
HighProbability HighProbability estimates which alternative
hypotheses have frequentist or
Bayesian probabilities
MSBVAR Bayesian Vector
Autoregression Models, Impulse Responses and Forecasting.
survBayes Fits a
proportional hazards model to time to event data by a Bayesian approach
tgp
Bayesian treed Gaussian process models
vabayelMix Variational Bayesian
Mixture Modelling
BsMD
Bayes Screening and Model Discrimination
evdbayes
Bayesian Analysis in Extreme Value Theory
siggenes
SAM and Efron's empirical Bayes approaches
mcmc
Markov Chain Monte Carlo
MCMCpack Markov chain Monte Carlo (MCMC) Package
A Few Good Books on Bayesian Analysis
I'll end this column by listing a few of my favorite books on
Bayesian Analysis:
Bayesian Statistics and Marketing
Applied Bayesian Modeling
Bayesian Approaches to Clinical Trials and Health-Care Evaluation
Biostatistics: A Bayesian Introduction
Bayesian Models for Categorical Data
Introduction to Bayesian Statistics
Bayesian Data Analysis
Bayesian Methods: A Social and Behavioral Sciences Approach
Return to top
|
