Content: The script main_script.R includes code to run a regression discontinuity (RD) design and validation and falsification of estimated results The folder data contains two files: bands_2016_2019.csv: a dataset of performers with additional information for each one. festivals_2016_2019.csv: a dataset of video search activity (as retrieved from Google Trends) for performers in file bands_2016_2019.csv The folder source contains two additional R scripts: data_preparation.R: generates the long dataset used to estimate RD effects status_simulation.R: randomly assigns treattment status to performers and estimates RD effects. Note this may take a long time to run. Parallel code is used: the number of cores has been set to 4. The folder simulation_results contains simulated data after running the script status_simulation.R.

R libraries required: doParallel foreach lubridate rdrobust rddensity rdlocrand stringr tidyverse