Plot an operating characteristic against the type I error rate
Source:R/plot_operating_characteristics_vs_tie.R
operating_characteristic_vs_tie.RdDraws one point per method and parameter combination, placing the operating characteristic against the type I error rate that method incurs, so that methods can be compared at the error rate they actually spend rather than at their nominal one.
Usage
operating_characteristic_vs_tie(
results_metrics_df,
case_study,
target_sample_size_per_arm,
treatment_effect,
operating_characteristic,
source_denominator_change_factor,
target_to_source_std_ratio,
show_tie_error_bars = FALSE
)Arguments
- results_metrics_df
The dataframe containing the results and metrics.
- case_study
The case study name.
- target_sample_size_per_arm
The target sample size per arm.
- treatment_effect
The treatment effect scenario ("consistent", "no_effect" or "partially_consistent").
- operating_characteristic
The metric to plot, as an entry of
frequentist_metricsorinference_metrics.- source_denominator_change_factor
The source denominator change factor.
- target_to_source_std_ratio
The target to source standard deviation ratio.
- show_tie_error_bars
Whether to draw the Monte Carlo interval on the type I error axis.