1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302
|
#' @title Parameters from Bayesian Models
#' @name model_parameters.brmsfit
#'
#' @description
#' Model parameters from Bayesian models. This function internally calls
#' [`bayestestR::describe_posterior()`] to get the relevant information for
#' the output.
#'
#' @param model Bayesian model (including SEM from **blavaan**. May also be
#' a data frame with posterior samples, however, `as_draws` must be set to
#' `TRUE` (else, for data frames `NULL` is returned).
#' @param ci Credible Interval (CI) level. Default to `0.95` (`95%`). See
#' [bayestestR::ci()] for further details.
#' @param group_level Logical, for multilevel models (i.e. models with random
#' effects) and when `effects = "all"` or `effects = "random"`,
#' include the parameters for each group level from random effects. If
#' `group_level = FALSE` (the default), only information on SD and COR
#' are shown.
#' @param component Which type of parameters to return, such as parameters for the
#' conditional model, the zero-inflation part of the model, the dispersion
#' term, or other auxiliary parameters be returned? Applies to models with
#' zero-inflation and/or dispersion formula, or if parameters such as `sigma`
#' should be included. May be abbreviated. Note that the *conditional*
#' component is also called *count* or *mean* component, depending on the
#' model. There are three convenient shortcuts: `component = "all"` returns
#' all possible parameters. If `component = "location"`, location parameters
#' such as `conditional`, `zero_inflated`, or `smooth_terms`, are returned
#' (everything that are fixed or random effects - depending on the `effects`
#' argument - but no auxiliary parameters). For `component = "distributional"`
#' (or `"auxiliary"`), components like `sigma`, `dispersion`, or `beta`
#' (and other auxiliary parameters) are returned.
#' @param as_draws Logical, if `TRUE` and `model` is of class `data.frame`,
#' the data frame is treated as posterior samples and handled similar to
#' Bayesian models. All arguments in `...` are passed to
#' `model_parameters.draws()`.
#' @inheritParams model_parameters.default
#' @inheritParams bayestestR::describe_posterior
#' @inheritParams insight::get_parameters
#'
#' @seealso [insight::standardize_names()] to rename columns into a consistent,
#' standardized naming scheme.
#'
#' @note When `standardize = "refit"`, columns `diagnostic`, `bf_prior` and
#' `priors` refer to the *original* `model`. If `model` is a data frame,
#' arguments `diagnostic`, `bf_prior` and `priors` are ignored.
#'
#' There is also a
#' [`plot()`-method](https://easystats.github.io/see/articles/parameters.html)
#' implemented in the [**see**-package](https://easystats.github.io/see/).
#'
#' @inheritSection model_parameters Confidence intervals and approximation of degrees of freedom
#'
#' @inheritSection model_parameters.zcpglm Model components
#'
#' @examplesIf require("rstanarm")
#' \donttest{
#' library(parameters)
#' model <- suppressWarnings(stan_glm(
#' Sepal.Length ~ Petal.Length * Species,
#' data = iris, iter = 500, refresh = 0
#' ))
#' model_parameters(model)
#' }
#' @return A data frame of indices related to the model's parameters.
#' @export
model_parameters.brmsfit <- function(model,
centrality = "median",
dispersion = FALSE,
ci = 0.95,
ci_method = "eti",
test = "pd",
rope_range = "default",
rope_ci = 0.95,
bf_prior = NULL,
diagnostic = c("ESS", "Rhat"),
priors = FALSE,
effects = "fixed",
component = "all",
exponentiate = FALSE,
standardize = NULL,
group_level = FALSE,
keep = NULL,
drop = NULL,
verbose = TRUE,
...) {
modelinfo <- insight::model_info(model, verbose = FALSE)
# Bayesian meta analysis
if (!insight::is_multivariate(model) && isTRUE(modelinfo$is_meta)) {
params <- .model_parameters_brms_meta(
model,
centrality = centrality,
dispersion = dispersion,
ci = ci,
ci_method = ci_method,
test = test,
rope_range = rope_range,
rope_ci = rope_ci,
diagnostic = diagnostic,
priors = priors,
exponentiate = exponentiate,
standardize = standardize,
keep_parameters = keep,
drop_parameters = drop,
...
)
} else if (effects %in% c("total", "random_total")) {
# group level total effects (coef())
params <- .group_level_total(model, centrality, dispersion, ci, ci_method, test, rope_range, rope_ci, ...)
params$Effects <- "total"
class(params) <- c("parameters_coef", "see_parameters_coef", class(params))
return(params)
} else {
# Processing
params <- .extract_parameters_bayesian(
model,
centrality = centrality,
dispersion = dispersion,
ci = ci,
ci_method = ci_method,
test = test,
rope_range = rope_range,
rope_ci = rope_ci,
bf_prior = bf_prior,
diagnostic = diagnostic,
priors = priors,
effects = effects,
component = component,
standardize = standardize,
keep_parameters = keep,
drop_parameters = drop,
verbose = verbose,
...
)
if (!(effects == "fixed" && component == "conditional")) {
random_effect_levels <- which(params$Effects == "random" & grepl("^(?!sd_|cor_)(.*)", params$Parameter, perl = TRUE) & !(params$Parameter %in% c("car", "sdcar")))
if (length(random_effect_levels) && isFALSE(group_level)) params <- params[-random_effect_levels, ]
}
# add prettified names as attribute. Furthermore, group column is added
params <- .add_pretty_names(params, model)
# exponentiate coefficients and SE/CI, if requested
params <- .exponentiate_parameters(params, model, exponentiate)
params <- .add_model_parameters_attributes(params,
model,
ci,
exponentiate,
ci_method = ci_method,
group_level = group_level,
verbose = verbose,
...
)
attr(params, "parameter_info") <- insight::clean_parameters(model)
attr(params, "object_name") <- insight::safe_deparse_symbol(substitute(model))
class(params) <- unique(c("parameters_model", "see_parameters_model", class(params)))
}
params
}
# brms meta analysis -------
.model_parameters_brms_meta <- function(model,
centrality = "median",
dispersion = FALSE,
ci = 0.95,
ci_method = "eti",
test = "pd",
rope_range = "default",
rope_ci = 0.95,
diagnostic = c("ESS", "Rhat"),
priors = FALSE,
exponentiate = FALSE,
standardize = NULL,
keep_parameters = NULL,
drop_parameters = NULL,
verbose = TRUE,
...) {
# parameters
smd <- insight::get_parameters(model, effects = "fixed", component = "conditional")
studies <- insight::get_parameters(model, effects = "random", parameters = "^(?!sd_)")
studies[] <- lapply(studies, function(i) i + smd[[1]])
tau <- insight::get_parameters(model, effects = "random", parameters = "^sd_")
params <- bayestestR::describe_posterior(
cbind(studies, smd),
centrality = centrality,
dispersion = dispersion,
ci = ci,
ci_method = ci_method,
test = test,
rope_range = rope_range,
rope_ci = rope_ci,
...
)
params_diagnostics <- bayestestR::diagnostic_posterior(
model,
effects = "all",
diagnostic = diagnostic,
...
)
params_tau <- bayestestR::describe_posterior(
tau,
centrality = centrality,
dispersion = dispersion,
ci = ci,
ci_method = ci_method,
test = test,
rope_range = rope_range,
rope_ci = rope_ci,
...
)
# add weights
params$Weight <- 1 / c(insight::get_response(model)[[2]], NA)
# merge description with diagnostic
params <- merge(params, params_diagnostics, by = "Parameter", all.x = TRUE, sort = FALSE)
# Renaming
re_name <- insight::find_random(model, flatten = TRUE)
study_names <- gsub(sprintf("r_%s\\[(.*)\\]", re_name[1]), "\\1", colnames(studies))
# replace dots by white space
study_names <- gsub(".", " ", study_names, fixed = TRUE)
# remove "Intercept"
study_names <- insight::trim_ws(gsub(",Intercept", "", study_names, fixed = TRUE))
cleaned_parameters <- c(study_names, "Overall", "tau")
# components
params$Component <- "Studies"
params_tau$Component <- "tau"
# merge with tau
params <- merge(params, params_tau, all = TRUE, sort = FALSE)
# reorder columns
ci_column <- which(colnames(params) == "CI_high")
weight_column <- which(colnames(params) == "Weight")
first_cols <- c(1:ci_column, weight_column)
params <- params[, c(first_cols, seq_len(ncol(params))[-first_cols])]
# filter parameters, if requested
if (!is.null(keep_parameters) || !is.null(drop_parameters)) {
params <- .filter_parameters(params,
keep = keep_parameters,
drop = drop_parameters,
verbose = verbose
)
}
# add attributes
attr(params, "tau") <- params_tau
attr(params, "pretty_names") <- cleaned_parameters
attr(params, "cleaned_parameters") <- cleaned_parameters
attr(params, "ci") <- ci
attr(params, "ci_method") <- ci_method
attr(params, "exponentiate") <- exponentiate
attr(params, "model_class") <- class(model)
attr(params, "is_bayes_meta") <- TRUE
attr(params, "study_weights") <- params$Weight
attr(params, "data") <- cbind(studies, smd, tau)
class(params) <- unique(c("parameters_brms_meta", "see_parameters_brms_meta", class(params)))
params
}
#' @export
standard_error.brmsfit <- function(model,
effects = "fixed",
component = "all",
...) {
effects <- insight::validate_argument(
effects,
c("fixed", "random")
)
component <- insight::validate_argument(
component,
c("all", "conditional", "zi", "zero_inflated")
)
params <- insight::get_parameters(model, effects = effects, component = component, ...)
.data_frame(
Parameter = colnames(params),
SE = unname(sapply(params, stats::sd, na.rm = TRUE))
)
}
#' @export
p_value.brmsfit <- p_value.BFBayesFactor
|