hib_fit_brms: Fit a brms style mixed-effects model

This function fits a brms model using the supplied database, formula and other function arguments tuning the model fitting process.

Usage

hib_fit_brms(
  obj,
  form,
  family = stats::gaussian,
  weight_specs = list(weights = NULL, weight_type = NULL),
  scale_specs = list(covars_to_scale = NULL, scale_func = NULL),
  relevel_specs = c(NULL),
  ...
)

Arguments

obj

A HibDatabase object or a list of them.

form

A formula in the style of brms (as in lme4) - including fixed and random effects.

family

A family function, e.g. gaussian, poisson - as allowed by brms

weight_specs

A named list containing "weights" and "weight_type".

weights: a numeric vector containing the relative weights to assign to each dataset/site, should sum to 1.
weight_type: a character, specifying either "dataset" or "site" - i.e. should the weights argument assign relative weights to whole datasets, or for sites in each dataset.

Only applicable if a list of HibDatabases are provided.
The length of 'weights' should be equal to the number of HibDatabases.
Note: if using this argument, the supplied formula should have a response of the form y | weights(data_weight) ~. The data_weight variable will be generated within the function call.

scale_specs

A named list containing "covars_to_scale" and "scale_func".

covars_to_scale is a character vector of numeric covariates that should be scaled prior to modelling.
scale_func is a function that will be used to scale the specified covariates - if NULL, base::scale() will be used.

relevel_specs

A vector where each item is: "<covar_to_relevel_name>" = "<base_level>".

...

Any other named arguments that can be passed to brms::brm(), e.g., prior, data2

Value

HibModelBRMS

Details

This function takes the supplied database object (and various other arguments), fits the specified model and returns a HibModelBRMS object.

References

Paul-Christian Bürkner (2017). brms: An R Package for Bayesian Multilevel Models Using Stan. Journal of Statistical Software, 80(1), 1-28. doi:10.18637/jss.v080.i01
Paul-Christian Bürkner (2018). Advanced Bayesian Multilevel Modeling with the R Package brms. The R Journal, 10(1), 395-411. doi:10.32614/RJ-2018-017
Paul-Christian Bürkner (2021). Bayesian Item Response Modeling in R with brms and Stan. Journal of Statistical Software, 100(5), 1-54. doi:10.18637/jss.v100.i05

Examples

if (FALSE) { # \dontrun{
  mod <- hib_fit_brms(obj, y ~ x + ( 1 | a ), ...)
} # }