Outlier detection in clinical registries - simulation study data and Stata code

Full description

Contains the simulated data and Stata code used to produce the results for the manuscript titled "Evaluating methods of outlier detection when benchmarking clinical registry data – a simulation study", accepted for publication in the Health Services and Outcomes Research Methodology Journal.

data_files.zip (code to generate all files in "do_files\simstudy1_preparation.do"):
raw_data - the .dta files produced from running the user-written hiersim command (https://doi.org/10.26180/24480889.v1)
summary_data - the .dta files produced from summarising of the results across each unique simulated scenario and method combination (performance measure average and 95% Monte Carlo confidence intervals)
parameter_check - the .dta files produced from summarising the simulated data parameters across each unique simulated scenario (performance measure average and 95% Monte Carlo confidence intervals)

do_files.zip:
simstudy1_preparation.do - the code to run the simulations (using the hiersim command, available at https://doi.org/10.26180/24480889.v1) and create summary datasets (performance measures and parameter checks)
simstudy1_manuscript.do - the code to produce the figures included in the main manuscript
simstudy1_supplementary.do - the code to produce the table and figures included in the manuscript supplementary material