Select a subset of the candidate hyperparameter settings,
and return a new varbvs analysis object with these hyperparameter
settings only.

# S3 method for varbvs
subset(x, subset, …)

## Arguments

x |
Output of function `varbvs` . |

subset |
Expression indicating hyperparameter settings to select.
This expression should include one or more of `logodds` ,
`sigma` and `sa` . |

… |
Other arguments to generic subset function. These
extra arguments are not used here. |

## Value

An object with S3 class `c("varbvs","list")`

.

## 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`

## Examples

# First run one of the examples in help(varbvs), then try running
# this.
#
# fit.new <- subset(fit,logodds < (-2))
#