---
license: gpl
language:
- en
tags:
- parquet
- securitiesandexchange
- financial
- balancesheet
- incomestatement
- cashflowstatement
- sec.gov
---
[TRY DATA HERE](https://www.kaggle.com/code/denytran/hugging-face-fsnonotesdatasets-starter/notebook)
OR
[UNZIPPED SAMPLE HERE](https://huggingface.co/datasets/DenyTranDFW/SEC_10K_FSNoNDS_Eat)
OR
[CREATE DATASETS](https://www.kaggle.com/code/denytran/tabular-financial-data-generator)
FINANCIAL STATEMENTS AND NO NOTES DATA SETS - 10-K FILINGS
SECTIONED BY [STANDARD INDUSTRIAL CLASSIFICATION](https://www.sec.gov/search-filings/standard-industrial-classification-sic-code-list)
- INDUSTRIAL_APPLICATIONS_AND_SERVICES
- OFFICE_OF_CRYPTO_ASSETS
- OFFICE_OF_ENERGY_AND_TRANSPORTATION
- OFFICE_OF_FINANCE
- OFFICE_OF_FINANCE_OR_OFFICE_OF_CRYPTO_ASSETS
- OFFICE_OF_LIFE_SCIENCES
- OFFICE_OF_MANUFACTURING
- OFFICE_OF_REAL_ESTATE_AND_CONSTRUCTION
- OFFICE_OF_TECHNOLOGY
- OFFICE_OF_TRADE_AND_SERVICES
**THIS REPOSITORY WILL CONTAIN THE FOLLOWING 10-K FINANCIAL STATEMENTS:**
[CURRENTLY, ONLY AVAILABLE UP TO 2023 WITH NO PLANS FOR UPDATES AS OF THIS TIME]
- INCOME STATEMENT
- CASH FLOW STATEMENT
- BALANCE SHEET
**SOURCE**:
- THIS DATA WAS PARSED FROM THE SEC'S [FINANCIAL STATEMENTS AND NOTES DATA SETS](https://www.sec.gov/dera/data/financial-statement-and-notes-data-set.html) AS A PERSONAL PROJECT
- IF YOU WANT THE ENTIRE DATASET (CREATED IN SQL SERVER), THE PARQUET FILES CAN BE FOUND [HERE](https://huggingface.co/datasets/DenyTranDFW/SEC-Financial-Statements-And-Notes-Dataset) ALONG WITH THE SQL SCRIPTS TO REBUILD THE DATABASE
**HOW THE FOLDERS ARE STRUCTURED**:
- \/\/\<10K_[CF | IS | BS]_CIK_ADSH_FYFYE.parquet>
WHERE:
- [CF | IS | BS] = STATEMENT (CASH FLOW STATEMENT, INCOME STATEMENT, BALANCE SHEET)
- CIK = CENTRAL INDEX KEY
- COMPANYNAME = NAME OF COMPANY
- ADSH = ACCESSION NUMBER
- FY & YEAR = FISCAL YEAR (YYYY)
- 10K = ANNUAL FINANCIAL REPORT (MORE INFO [HERE](https://www.sec.gov/files/form10-k.pdf))
- FYE = FISCAL YEAR END (MMDD)
**REGARDING COMPANY NAME**:
IF A COMPANY HAS CHANGED IT'S NAME, THE ENTIRE DIRECTORY WAS MOVED INTO THE DIRECTORY WITH THE LATEST YEAR, FOR EXAMPLE, THE CIK ID OF 0000077360 HAS:
- 0000077360_PENTAIR_INC
- 2010
- 2011
- 0000077360_PENTAIR_LTD
- 2012
- 2013
- 0000077360_PENTAIR_PLC
- 2014
- ...
- 20XX
- ...
- 2023
THE DIRECTORY STRUCTURE WOULD THEN BECOME:
- 0000077360_PENTAIR_PLC
- 2014
- ...
- 20XX
- ...
- 2023
- 0000077360_PENTAIR_INC
- 2010
- 2011
- 0000077360_PENTAIR_LTD
- 2012
- 2013
**ADDITIONAL NOTES**:
- THERE ARE SOME FILES MISSING FROM THIS, WHICH I WOULD NEED TO TROUBLESHOOT TO DETERMINE WHY (THOSE FILES WILL BE UPLOADED AT A LATER TIME).
- IF A SINGLE STATEMENT DID NOT PROVIDE AN OUTPUT, ALL THREE WILL BE EXCLUDED UNTIL THE CAUSE IS DETERMINED
- THE DATA WAS EXPORTED VIA A SCRIPT FROM MSSQL, USING PYTHON, WITH VERY LITTLE ERROR CHECKING, SO ASSUME ERRORS WILL BE PRESENT
- FORGIVE ME IF THERE ARE OUTLIERS TO THIS, AS SOME OF THE SORTING ARE DONE MANUALLY
- WITH THAT SAID, THIS DATA SHOULD BE TREATED AS, MORE-OR-LESS, A DUMMY DATA SET