Skip to content

ExecuComp regression data

This analysis dataset joins matched executive compensation to HCT match quality and annual Compustat controls.

flowchart TB
    pay[(Matched executive pay)]
    scores[(Worker-firm-period scores)]
    comp[(Processed annual Compustat)]
    build[Join executive-firm-years]
    data[(ExecuComp regression data)]

    pay --> build
    scores --> build
    comp --> build
    build --> data

    classDef source fill:#e8eef7,stroke:#52739e
    classDef process fill:#fff0cc,stroke:#a66b00
    classDef data fill:#e2f2e7,stroke:#3b7a4b
    class pay,scores,comp source
    class build process
    class data data
Inputs
Dataset Location
Matched executive pay /data/full/raw/execucomp/execucomp_revelio_matched_pay.parquet
Worker-firm-period scores /data/<coverage>/analyses/<mode>/worker_firm_period_scores/
Processed annual Compustat /data/full/processed/compustat_annual.parquet
Build
Component Location
Builder analyses/modules/disaggregated_execucomp.R
Runtime hct-tower-analyses-cpu
Entry point analyses/pipelines/disaggregated_execucomp.R / build-execucomp-regression-table
Output
Dataset File / location Grain Key and variables
ExecuComp regression data execucomp_regression_table.parquet in /data/<coverage>/analyses/<mode>/ Executive-firm-year Key: execid, gvkey, user_id, company_id, calendar_year
Variables include match_score, hazard_prob, log_wage, salary, total_compensation, at, sale

For each executive-firm-year, the builder retains the latest matching scored worker-firm period and attaches Compustat fields by gvkey and year. Separate files are built for training and full_training scores.

Rscript --vanilla pipelines/disaggregated_execucomp.R build_regression_table \
  --config config/config_full.yaml --mode training --overwrite
Rscript --vanilla pipelines/disaggregated_execucomp.R build_regression_table \
  --config config/config_full.yaml --mode full_training --overwrite
slurm/submit.sh --task build-execucomp-regression-table \
  --mode training --config config_full.yaml --overwrite
slurm/submit.sh --task build-execucomp-regression-table \
  --mode full_training --config config_full.yaml --overwrite

The output is consumed by the executive-compensation analysis.