Skip to content

Public-firm performance and innovation data

This analysis pipeline combines HCT firm-year measures with accounting, returns, and patent outcomes for the firm-performance and innovation tests.

Blue nodes are upstream inputs, amber nodes are local support datasets, green nodes are reusable analysis-layer datasets, and the purple node is the final dataset read by analysis code. The dashed amber node is not wired into this build.

flowchart TB
    scores[(Worker-firm-period scores)]
    company_mq[(Company-year MQ)]
    mapping[(Revelio-Compustat-PERMNO mapping)]
    patents_raw[(Raw KPSS patents)]
    returns_raw[(Raw CRSP monthly returns)]
    controls[(Compustat analysis controls)]

    mq_stats[(Firm-year MQ statistics)]
    patent_panel[(Analysis patent panel)]
    volatility[(Returns volatility)]
    rnd_mq[(R&D-worker MQ panel)]
    final[(Public-firm analysis dataset)]

    scores --> mq_stats
    scores -.-> rnd_mq
    patents_raw --> patent_panel
    mapping --> patent_panel
    returns_raw --> volatility

    company_mq --> final
    mq_stats --> final
    patent_panel --> final
    volatility --> final
    mapping --> final
    controls --> final

    classDef input fill:#e8eef7,stroke:#52739e,stroke-width:2px
    classDef support fill:#fff0cc,stroke:#a66b00,stroke-width:2px
    classDef reusable fill:#e2f2e7,stroke:#3b7a4b,stroke-width:2px
    classDef analysis fill:#eee5f5,stroke:#76528c,stroke-width:2px
    classDef dormant fill:#fff0cc,stroke:#a66b00,stroke-width:2px,stroke-dasharray:5 5
    class scores,mapping,patents_raw,returns_raw,controls input
    class mq_stats,patent_panel,volatility support
    class company_mq reusable
    class final analysis
    class rnd_mq dormant

Analysis-support datasets

The public-firm analysis dataset joins these products to company-year MQ and Compustat controls. It is the terminal analysis-specific dataset produced by this branch.

The separate company-year MQ and firm-year MQ-statistics branches both aggregate the same score data. See the firm-year overview for the overlapping columns and current duplication.

rnd_mq_panel.parquet remains configured in the public-firm YAML files and a builder exists in analyses/modules/build_rnd_mq_panel.R, but no current data entry point invokes it and it does not feed table89_dataset.parquet.