metabias-class {metabias}

metabias S3 class

Description

A object of class metabias is the result of fitting one or more models to a dataset with one row per study being meta-analyzed. These models are either (1) a meta-analysis with a correction for one or more within-study or across-study biases, or (2) a sensitivity analysis for meta-analyses with respect to these biases. Examples of functions that return such objects include:

• PublicationBias::pubbias_meta()

• PublicationBias::pubbias_svalue()

• phacking::phacking_meta()

• multibiasmeta::multibias_meta()

• multibiasmeta::multibias_evalue()

Usage

metabias(
  data = data.frame(),
  values = list(),
  stats = data.frame(),
  fits = list()
)

new_metabias(x = list())

## S3 method for class 'metabias'
summary(object, ...)

Arguments

data

Dataframe containing data used to fit the model(s), with added columns for any values computed during model fitting.

values

List of values of arguments passed to the function.

stats

Dataframe of summary statistics from the model fit(s).

fits

List of fitted objects (which have a class that depends on the underlying fitting methods, e.g. robumeta::robu or rstan::stanfit).

x

List with elements "data", "values", "stats", "fits".

object

Object of class metabias.

...

Not used.

Value

An object of class metabias, which consists of a list containing the elements data, values, stats, fits (corresponding to the arguments passed).

Examples

# example model from robumeta::robu()
hier_mod <- robumeta::robu(effectsize ~ binge + followup + sreport + age,
                           data = robumeta::hierdat, studynum = studyid,
                           var.eff.size = var, modelweights = "HIER",
                           small = TRUE)

ci <- 0.95  # example set value
hier_mb <- metabias(data = robumeta::hierdat,                 # data passed to model
                    values = list(ci_level = ci),             # value used
                    stats = robu_ci(hier_mod, ci_level = ci), # stats from model
                    fits = list("robu" = hier_mod))           # model object

hier_mb
summary(hier_mb)