BayesReg: Bayesian linear regression output In bayess: Bayesian Essentials with R

Description

This function contains the R code for the implementation of Zellner's G-prior analysis of the regression model as described in Chapter 3. The purpose of `BayesRef` is dual: first, this R function shows how easily automated this approach can be. Second, it also illustrates how it is possible to get exactly the same type of output as the standard R function `summary(lm(y~X))`. In particular, it calculates the Bayes factors for variable selection, more precisely single variable exclusion.

Usage

 `1` ```BayesReg(y, X, g = length(y), betatilde = rep(0, dim(X)[2]), prt = TRUE) ```

Arguments

 `y` response variable `X` matrix of regressors `g` constant g for the G-prior `betatilde` prior mean on beta `prt` boolean variable for printing out the standard output

Value

 `postmeancoeff ` posterior mean of the regression coefficients `postsqrtcoeff ` posterior standard deviation of the regression coefficients `log10bf ` log-Bayes factors against the full model `postmeansigma2` posterior mean of the variance of the model `postvarsigma2` posterior variance of the variance of the model

Examples

 ```1 2``` ```data(faithful) BayesReg(faithful[,1],faithful[,2]) ```

Example output

```Loading required package: MASS

Attaching package: 'gplots'

The following object is masked from 'package:stats':

lowess

Attaching package: 'combinat'

The following object is masked from 'package:utils':

combn

PostMean PostStError Log10bf EvidAgaH0
Intercept   3.4878      0.0304
x1          1.0225      0.0303     Inf    (****)

Posterior Mean of Sigma2: 0.2513
Posterior StError of Sigma2: 0.3561

\$postmeancoeff
[1] 3.487783 1.022509

\$postsqrtcoeff
[1] 0.03039825 0.03034252

\$log10bf
[,1]
[1,]  Inf

\$postmeansigma2
[1] 0.2513425

\$postvarsigma2
[1] 0.1268176
```

bayess documentation built on May 29, 2017, 9:39 p.m.