Skip to contents

Resamples complete independent study or outer-cluster blocks and refits MARS. Duplicate copies receive fresh study, random-group, and robust-cluster IDs.

Usage

bootstrap_mars(
  data,
  studyID,
  effectID = NULL,
  sample_size = NULL,
  effectsize_type = NULL,
  formula = NULL,
  variable_names = NULL,
  effectsize_name = NULL,
  estimation_method = NULL,
  variance = NULL,
  varcov_type = NULL,
  weights = NULL,
  structure = NULL,
  intercept = FALSE,
  missing = "remove",
  optim_method = "L-BFGS-B",
  robustID = NULL,
  multivariate_covs = NULL,
  lasso = FALSE,
  lasso_args = list(),
  tol = NULL,
  ...
)

Arguments

data

Data used for analysis

studyID

Character string representing the study ID

effectID

Character string representing the effect size ID

sample_size

Character string representing the sample size of the studies.

effectsize_type

Type of effect size being analyzed

formula

The formula used for specifying the fixed and random structure. Used for univariate and multilevel structures.

variable_names

Vector of character strings representing the attributes with correlations. The attributes that are correlated should be separated by an underscore.

effectsize_name

Character string representing the name of the effect size column in the data.

estimation_method

Estimator. NULL uses REML for ordinary models and MLE for LASSO. Explicit LASSO estimation must use MLE.

variance

Character string representing the name of the variance of the effect size in the data.

varcov_type

Sampling covariance type. NULL uses the formula structure or weighted correlation covariance. Specify the SMD covariance type explicitly.

weights

User specified matrix of weights for analysis.

structure

Between-study covariance structure. NULL infers univariate or multilevel from formula; effect-size inputs use UN, or multivariate for LASSO.

intercept

Whether a model intercept should be specified, default is FALSE meaning no intercept. See details for more information.

missing

Whether missing data should be removed, or kept. Default is removing.

optim_method

Optimization method that is passed to the optim function. Default is 'L-BFGS-B'.

robustID

A character vector specifying the cluster group to use for computing the robust standard errors.

multivariate_covs

A one-sided formula to specify the covariates used in a multivariate analysis.

lasso

Fit the current mars_lasso() engine in every sample. Summaries describe coefficient and selection stability without inferential intervals.

lasso_args

Named controls for mars_lasso(), including lambda_grid, folds, selection, seed, maxit and tol. Cross-validation preprocessing is learned separately within each training fold.

tol

NULL retains the fitting engine's default: 1e7 factr for ordinary L-BFGS-B fits, or 1e-7 numerical tolerance for LASSO. Supply a LASSO tolerance here or in lasso_args, once.

...

Additional arguments forwarded to mars(), such as tau2 or robust_args.

Examples

# \donttest{
boot_fit <- bootstrap_mars(
  data = teacher_expectancy,
  studyID = "study",
  effectID = NULL,
  sample_size = NULL,
  formula = yi ~ 1,
  variance = "vi",
  varcov_type = "univariate",
  structure = "univariate"
)
# }