run_propensity_matching: Performs propensity score matching on a data.table and gives...
In zbeaver4/beavR: Useful and Reusable R Functions from Past Projects

View source: R/propensity_matching_functions.R

Given a data.table with information about which entities should be matched against, performs propensity score matching using nearest neighbors and saves out diagnostic plots. Returns a list of those entities which are significantly different from the population

run_propensity_matching(dt, cols_of_interest, reg_outcome_var, match_exclude_cols, bad_entities, entity_name, plot_fp, list_save_fp, list_name, control_match_num = 3, verbose = T, sample_size = NULL, entity_sample = NULL)

`dt:`	data.table containin all the data
`cols_of_interest:`	columns on which matching should occur
`reg_outcome_var:`	character indicating the outcome column variable
`match_exclude_cols:`	character vector of columns to exclude when performing the matching
`bad_entities:`	a character vector indicating the entities for which propensity score matching should be performed
`entity_name:`	character vector of the column in dt to be used subsetting down and identifiying 'bad_entities'
`plot_fp:`	character filepath indicating where the psm matching plots should be stored
`list_save_fp:`	character filepath indicating where the list of significant entities should be stored
`control_match_num:`	The number of properties to be matched from the global population
`verbose:`	Boolean indicating whether or not to give progress updates
`sample_size:`	numeric indicating the sample size of the entity of interest to be taken for matching