This is a function for computing the information level from a dataset. For
multi-stage designs, earlier results can be passed as a
monitored_design object.
Usage
estimate_information(
data,
monitored_design,
estimation_function,
estimation_arguments,
correction_function = NULL,
orthogonalize = NULL,
rng_seed,
return_results = FALSE,
control = monitored_analysis_control()
)Arguments
- data
A data.frame containing the data to be analyzed. A column named
.idindicates which observations correspond to each individual.- monitored_design
An object of class
monitored_designcreated usinginitialize_monitored_design()- estimation_function
A function whose arguments include a data.frame named
data- estimation_arguments
A list of any additional arguments needed by
estimation_function- correction_function
A function which takes named arguments from
estimation_argumentsand returns a numeric scalar: this is used to scale the variance before computing information.- orthogonalize
Logical scalar: Should estimates, their covariance, and the resulting test statistics be orthogonalized?
- rng_seed
Numeric scalar containing the L'Ecuyer pseudorandom number generator seed
- return_results
Logical scalar: Should estimates be returned, or only the covariance and information levels?
- control
A list containing the control arguments for computation, typically created with
monitored_analysis_control()
See also
calculate_covariance() for computing the covariance matrix
of estimators across analyses, monitored_analysis_control for
details about the default computing arguments, monitored_analysis
for conducting analyses at pre-specified information levels