Aggregates treatment effects across multiple outcomes, time periods, or subgroups with appropriate statistical adjustments for multiple testing.
Usage
aggregate_treatment_effects(
estimation_results,
aggregation_method = "weighted_average",
weight_var = "inv_var",
multiple_testing_correction = "holm",
confidence_level = 0.95
)Arguments
- estimation_results
List of estimation results from DiD, event study, or synthetic control
- aggregation_method
Character. Method: "simple_average", "weighted_average", "meta_analysis". Default: "weighted_average"
- weight_var
Character. Variable to use for weighting (e.g., "n_obs", "inv_var"). Default: "inv_var"
- multiple_testing_correction
Character. Method: "bonferroni", "holm", "fdr". Default: "holm"
- confidence_level
Numeric. Confidence level for intervals. Default: 0.95
Value
A list containing:
- aggregated_effects
Summary of aggregated treatment effects
- individual_effects
Individual effect estimates with corrections
- heterogeneity_tests
Tests for effect heterogeneity
- meta_analysis_results
Meta-analysis results if applicable
Examples
if (FALSE) { # \dontrun{
aggregated_results <- aggregate_treatment_effects(
estimation_results = list(did_results, event_study_results),
aggregation_method = "meta_analysis",
multiple_testing_correction = "fdr"
)
} # }