Dataset

credit_card_transactions

Card & Transactions Download 1 source

credit_card_transactions is a small tabular dataset published by aegisheld containing anonymized customer-level credit card account and spending summaries. It is distributed as CSV with a single training split of 8,950 rows.

Data preview

Loading preview...

Dataset card

What this dataset contains

Each row represents a single cardholder identified by a string CUST_ID and summarized behavioral aggregates rather than individual transaction records. Coverage is tagged as US, and the overall size band is 1K to 10K rows, consistent with the 8,950-row train split. The license field is unstated on the source page, even though the dataset carries a license:mit tag in its metadata.

Structure and fields

The schema has 18 columns. Identifier CUST_ID is a string, while account-level numerics include BALANCE, CREDIT_LIMIT, PAYMENTS, MINIMUM_PAYMENTS, and TENURE (int64). Spending behavior is captured by PURCHASES together with ONEOFF_PURCHASES, INSTALLMENTS_PURCHASES, and CASH_ADVANCE, plus their companion frequency ratios (PURCHASES_FREQUENCY, ONEOFF_PURCHASES_FREQUENCY, PURCHASES_INSTALLMENTS_FREQUENCY, CASH_ADVANCE_FREQUENCY) and transaction counts CASH_ADVANCE_TRX and PURCHASES_TRX, with PRC_FULL_PAYMENT indicating propensity to pay the full statement. The file is delivered as CSV and is readily loaded with pandas, polars, or the datasets and mlcroissant libraries.

How analysts use it

Because features are already aggregated per customer, the file is commonly used for retail-banking segmentation, credit-utilization profiling, and benchmark exercises on purchase-versus-cash-advance mix. Practitioners can derive signals such as installment share (INSTALLMENTS_PURCHASES divided by PURCHASES) or revolving behavior from PRC_FULL_PAYMENT, then map those to card-portfolio cohort views. Given the US tag and modest size, the data is best treated as a reference sample for methodology prototyping rather than a production-scale transaction feed.

Documentation & references 1
  1. aegishield/credit_card_transactions: listing on Hugging Face

Further reading

All guides
  • What is alternative data?Alternative data in investment research: what the term covers, how it differs from data format, and why provenance, timing, access and validation matter.
  • Card transaction panelsHow anonymized card panels turn payment flows into revenue estimates: panel construction, debiasing, aggregation lags, and the gap between what panels see and what companies report.
  • Alternative data for investingHow funds turn alternative data into an investment edge: signal families, evaluation, backtesting, compliance and the build-vs-buy decision, mapped to the datasets and providers indexed on this encyclopedia.

Discussion

Questions or observations about this dataset? Join the conversation on X.

Discuss on X

No comments yet. Start the conversation on X.