summary.varbvs: Summarize a fitted variable selection model.

In varbvs: Large-Scale Bayesian Variable Selection Using Variational Methods

Summarize a fitted variable selection model.

Description

Generate a summary of the Bayesian variable selection model fitted using variational approximation methods.

Usage

  ## S3 method for class 'varbvs'
summary(object, cred.int = 0.95, nv, pip.cutoff, ...)
  ## S3 method for class 'summary.varbvs'
print(x, digits = 3, ...)
  ## S3 method for class 'varbvs'
print(x, digits = 3, ...)

Arguments

object	Output of function varbvs.
cred.int	Size of credible interval, a number between 0 and 1.
nv	Show detailed statistics for top nv variables, ranked according to their posterior inclusion probabilities. Only one of nv and pip.cutoff may be specified. If neither are specified, the default is nv = 5.
pip.cutoff	Show detailed statistics for all variables in which the posterior inclusion probability (PIP) is at least pip.cutoff. Only one of nv and pip.cutoff may be specified.
x	Output of function summary.varbvs.
digits	Number of digits shown when printing posterior probabilities of top nv variables.
...	Additional summary or print arguments.

Details

The printed summary is divided into three parts. The first part summarizes the data and optimization settings. It also reports the hyperparameter setting that yields the largest marginal likelihood—more precisely, the approximate marginal likelihood computed using the variational method. For the linear regression only (family = "gaussian") when no additional covariates (Z) are included, it reports the estimated proportion of variance in the outcome explained by the model (PVE), and the credible interval of the PVE estimate brackets.

The second part summarizes the approximate posterior distribution of the hyperparameters (sigma, sa, logodds). The "estimate" column is the value averaged over hyperparameter settings, treating objectlogw as (unnormalized) log-marginal probabilities. The next column, labeled "Pr>x", where x = cred.int gives the credible interval based on these weights (computed using function cred).

The third part summarizes the variable selection results. This includes the total number of variables included in the model at different posterior probability thresholds, and a more detailed summary of the variables included in the model with highest posterior probability. For family = "gaussian", the "PVE" column gives the estimated proportion of variance in the outcome explained by the variable (conditioned on being included in the model).

Value

An object of class summary.varbvs, to be printed by print.summary.varbvs.

Author(s)

Peter Carbonetto [e-mail]

References

P. Carbonetto and M. Stephens (2012). Scalable variational inference for Bayesian variable selection in regression, and its accuracy in genetic association studies. Bayesian Analysis 7, 73–108.

See Also

varbvs, varbvs.properties

Examples

  # See help(varbvs) for examples.

varbvs documentation built on June 7, 2023, 5:43 p.m.