unma.run: Run univariate network meta-analysis (NMA) anlaysis

In vandy10s/mnma: Multivariate Network Meta-Analysis (NMA) within a Bayesian Framework

Description

The function conducts univariate NMA using a mnma.model object as an input. It uses OpenBUGS software (through the R2OpenBUGS package).

Usage

unma.run(input, inits = NULL, n.iter = 5000, n.burnin = 2000,
  n.chains = 1, n.thin = 1, debug = FALSE, codaPkg = FALSE)

Arguments

input	mnma.model class data, which can be generated using mnma.model function
inits	a list of initial values for OpenBUGS model with n.chains elements; If inits=NULL (default), initial values are randomly generated by OpenBUGS.
n.iter	number of total iterations per chain (default: 5000)
n.burnin	length of burnin per chain (default: 2000)
n.chains	number of Markov chains (default: 1)
n.thin	thining (default: 1)
debug	if FALSE (default), OpenBUGS is closed automatically after running the unma.run script. If TRUE, OpenBUGS remains open with additional information.
codaPkg	if FALSE (default), a bugs object is returned. Otherwise, file names of OpenBUGS output are returned with TRUE.

Value

mnma.result class bugs result. See bugs for details.

References

Lu and Ades (2006) Journal of the American Statistical Association 101(474): 447-459 (doi: 10.1198/016214505000001302)

Dias et al. (2010) Statistics in Medicine 29(7-8): 932-944 (doi: 10.1002/sim.3767)

See Also

bugs

Examples

## Not run: 
# Run after data transformation using mnma.model function.
res <- unma.run(data, n.iter=3000, n.burnin=1000)
names(res$outcome1)
# provides mean, standard deviation, and percentile information about log odds ratio
res$outcome1$summary
res$outcome2$summary

## End(Not run)

vandy10s/mnma documentation built on May 19, 2019, 8:24 a.m.