XBRL filing data¶
The source data are SEC company-filing facts in XBRL format. The project uses
reprocessed long-form extracts of numeric and text facts from 10-K and 10-Q
filings. These are staged under /data/xbrl/ in the data-processing
environment; the extraction and reprocessing that created them are external to
this repository.
Available source files
| Input | Location | Role |
|---|---|---|
| 10-K numeric facts | /data/xbrl/form_10K_notes_numeric_long.parquet |
Numeric XBRL facts and filing contexts reported in annual filings |
| 10-Q numeric facts | /data/xbrl/form_10Q_notes_numeric_long.parquet |
Numeric XBRL facts and filing contexts reported in quarterly filings |
| 10-K text facts | /data/xbrl/form_10K_notes_text_long.parquet |
Text XBRL facts from annual filings |
| 10-Q text facts | /data/xbrl/form_10Q_notes_text_long.parquet |
Text XBRL facts from quarterly filings |
Legacy compatibility input
/data/xbrl_goodwill/goodwill_xbrl_20102022.csv is an older curated extract
(a repository copy remains at shared/data/external/goodwill_xbrl_20102022.csv).
It remains configured for compatibility and is unioned with the current
deal panel by the target and acquirer mapping builders. It is not the source
used to build the current deal panel for the Table 9 regression. Its
filename therefore does not establish the coverage of the current analysis
sample.
The source files contain filings across the years available in the staged extracts. Their current date coverage depends on those files and is not specified by a year range on this page. The repository does not contain a reproducible SEC downloader or the upstream process that generated these reprocessed extracts.
See processed deal data for how the repository uses these source files to build the clean deal panel and target/acquirer mappings.