RoBMA (version 3.5.0)

BiBMA: Estimate a Bayesian Model-Averaged Meta-Analysis of Binomial Data

Description

BiBMA estimate a binomial-normal Bayesian model-averaged meta-analysis. The interface allows a complete customization of the ensemble with different prior (or list of prior) distributions for each component.

Usage

BiBMA(
  x1,
  x2,
  n1,
  n2,
  study_names = NULL,
  study_ids = NULL,
  rescale_priors = 1,
  priors_effect = set_default_binomial_priors("effect", rescale = rescale_priors),
  priors_heterogeneity = set_default_binomial_priors("heterogeneity", rescale =
    rescale_priors),
  priors_effect_null = set_default_binomial_priors("effect", null = TRUE),
  priors_heterogeneity_null = set_default_binomial_priors("heterogeneity", null = TRUE),
  priors_hierarchical = set_default_binomial_priors("hierarchical"),
  priors_hierarchical_null = set_default_binomial_priors("hierarchical", null = TRUE),
  priors_baseline = set_default_binomial_priors("baseline"),
  priors_baseline_null = set_default_binomial_priors("baseline", null = TRUE),
  chains = 3,
  sample = 5000,
  burnin = 2000,
  adapt = 500,
  thin = 1,
  parallel = FALSE,
  autofit = TRUE,
  autofit_control = set_autofit_control(),
  convergence_checks = set_convergence_checks(),
  algorithm = "bridge",
  save = "all",
  seed = NULL,
  silent = TRUE,
  ...
)

Value

NoBMA returns an object of class 'RoBMA'.

Arguments

x1

a vector with the number of successes in the first group

x2

a vector with the number of successes in the second group

n1

a vector with the number of observations in the first group

n2

a vector with the number of observations in the second group

study_names

an optional argument with the names of the studies

study_ids

an optional argument specifying dependency between the studies (for using a multilevel model). Defaults to NULL for studies being independent.

rescale_priors

a re-scaling factor for the prior distributions. The re-scaling factor allows to adjust the width of all default priors simultaneously. Defaults to 1.

priors_effect

list of prior distributions for the effect size (mu) parameter that will be treated as belonging to the alternative hypothesis. Defaults to prior(distribution = "student", parameters = list(location = 0, scale = 0.58, df = 4)), based on logOR meta-analytic estimates from the Cochrane Database of Systematic Reviews bartos2023empiricalRoBMA.

priors_heterogeneity

list of prior distributions for the heterogeneity tau parameter that will be treated as belonging to the alternative hypothesis. Defaults to prior(distribution = "invgamma", parameters = list(shape = 1.77, scale = 0.55)) that is based on heterogeneities of logOR estimates from the Cochrane Database of Systematic Reviews bartos2023empiricalRoBMA.

priors_effect_null

list of prior distributions for the effect size (mu) parameter that will be treated as belonging to the null hypothesis. Defaults to a point null hypotheses at zero, prior(distribution = "point", parameters = list(location = 0)).

priors_heterogeneity_null

list of prior distributions for the heterogeneity tau parameter that will be treated as belonging to the null hypothesis. Defaults to a point null hypotheses at zero (a fixed effect meta-analytic models), prior(distribution = "point", parameters = list(location = 0)).

priors_hierarchical

list of prior distributions for the correlation of random effects (rho) parameter that will be treated as belonging to the alternative hypothesis. This setting allows users to fit a hierarchical (three-level) meta-analysis when study_ids are supplied. Note that this is an experimental feature and see News for more details. Defaults to a beta distribution prior(distribution = "beta", parameters = list(alpha = 1, beta = 1)).

priors_hierarchical_null

list of prior distributions for the correlation of random effects (rho) parameter that will be treated as belonging to the null hypothesis. Defaults to NULL.

priors_baseline

prior distributions for the alternative hypothesis about intercepts (pi) of each study. Defaults to NULL.

priors_baseline_null

prior distributions for the null hypothesis about intercepts (pi) for each study. Defaults to an independent uniform prior distribution for each intercept prior("beta", parameters = list(alpha = 1, beta = 1), contrast = "independent").

chains

a number of chains of the MCMC algorithm.

sample

a number of sampling iterations of the MCMC algorithm. Defaults to 5000.

burnin

a number of burnin iterations of the MCMC algorithm. Defaults to 2000.

adapt

a number of adaptation iterations of the MCMC algorithm. Defaults to 500.

thin

a thinning of the chains of the MCMC algorithm. Defaults to 1.

parallel

whether the individual models should be fitted in parallel. Defaults to FALSE. The implementation is not completely stable and might cause a connection error.

autofit

whether the model should be fitted until the convergence criteria (specified in autofit_control) are satisfied. Defaults to TRUE.

autofit_control

allows to pass autofit control settings with the set_autofit_control() function. See ?set_autofit_control for options and default settings.

convergence_checks

automatic convergence checks to assess the fitted models, passed with set_convergence_checks() function. See ?set_convergence_checks for options and default settings.

algorithm

a string specifying the algorithm used for the model averaging. Defaults to "bridge" which results in estimating individual models using JAGS and computing the marginal likelihood using bridge sampling. An alternative is "ss" which uses spike and slab like parameterization to approximate the Bayesian model averaging with a single model. Note that significantly more sample, burnin, and adapt iterations are needed for the "ss" algorithm.

save

whether all models posterior distributions should be kept after obtaining a model-averaged result. Defaults to "all" which does not remove anything. Set to "min" to significantly reduce the size of final object, however, some model diagnostics and further manipulation with the object will not be possible.

seed

a seed to be set before model fitting, marginal likelihood computation, and posterior mixing for reproducibility of results. Defaults to NULL - no seed is set.

silent

whether all print messages regarding the fitting process should be suppressed. Defaults to TRUE. Note that parallel = TRUE also suppresses all messages.

...

additional arguments.

Details

The BiBMA() function estimates the binomial-normal Bayesian model-averaged meta-analysis described in bartos2023empirical;textualRoBMA. See vignette("MedicineBiBMA", package = "RoBMA") vignette for a reproduction of the oduwole2018honey;textualRoBMA example. Also RoBMA() for additional details.

Generic summary.RoBMA(), print.RoBMA(), and plot.RoBMA() functions are provided to facilitate manipulation with the ensemble. A visual check of the individual model diagnostics can be obtained using the diagnostics() function. The fitted model can be further updated or modified by update.RoBMA() function.

References

See Also

RoBMA(), summary.RoBMA(), update.RoBMA(), check_setup()

Examples

if (FALSE) {
# using the example data from Oduwole (2018) and reproducing the example from
# Bartos et al. (2023) with domain specific informed prior distributions

fit <- BiBMA(
  x1          = c(5, 2),
  x2          = c(0, 0),
  n1          = c(35, 40),
  n2          = c(39, 40),
  priors_effect        = prior_informed(
      "Acute Respiratory Infections",
      type = "logOR", parameter = "effect"),
  priors_heterogeneity = prior_informed(
      "Acute Respiratory Infections",
      type = "logOR", parameter = "heterogeneity")
 )

 summary(fit)

 # produce summary on OR scale
 summary(fit, output_scale = "OR")

}