--- license: cc-by-4.0 language: - en pretty_name: CalcFi Open Data size_categories: - 100K **Live source + JSON API:** https://calcfi.app/developers > **Per-series HTML pages with charts:** https://calcfi.app/data > **OpenAPI 3.1 spec:** https://calcfi.app/api/insights/openapi.json This repo mirrors the read-only data exposed by [CalcFi](https://calcfi.app) so it can be consumed as a Frictionless Data Package, mirrored to dataset registries, and version-controlled with full git history. ## What's in here 34 series across 11 categories, each in its own subdir under [`datasets/`](datasets/): | Category | Series | |---|---| | Mortgage rates | 30-year-fixed, 15-year-fixed, 10-year-treasury | | Interest rates | federal-funds-rate, prime-rate, 2-year-treasury | | Inflation | cpi, pce | | Employment | unemployment-rate, labor-force-participation, avg-hourly-earnings | | Energy | crude-oil-wti, crude-oil-brent, us-regular-gas-price | | FX | usd-eur, usd-gbp, usd-jpy | | Crypto | bitcoin, ethereum, solana | | Deposit rates | savings-national-rate, money-market-national-rate, cd-3mo, cd-12mo, cd-60mo | | Credit | credit-card-apr, auto-loan-rate-48mo, personal-loan-rate-24mo | | Commodities | copper-price, corn-price | | Global (World Bank) | us-gdp-per-capita, world-inflation, world-unemployment, eurozone-gdp-per-capita | Every subdir contains: - `data.csv` — full history (date, value, unit), provenance headers preserved - `datapackage.json` — [Frictionless Data Package](https://frictionlessdata.io/) descriptor, DataHub-compatible - `README.md` — series-specific source, license, citation, and CalcFi live URL ## Quick start ```bash # Full mortgage rate history curl https://raw.githubusercontent.com/jerehere/calcfi-open-data/main/datasets/30-year-fixed/data.csv # CPI history curl https://raw.githubusercontent.com/jerehere/calcfi-open-data/main/datasets/cpi/data.csv # Latest single value (live API, daily-refreshed) curl https://calcfi.app/api/rates/30-year-fixed ``` ```python import pandas as pd url = "https://raw.githubusercontent.com/jerehere/calcfi-open-data/main/datasets/30-year-fixed/data.csv" df = pd.read_csv(url, comment="#", parse_dates=["date"]) print(df.tail()) ``` ```r library(readr) df <- read_csv("https://raw.githubusercontent.com/jerehere/calcfi-open-data/main/datasets/30-year-fixed/data.csv", comment = "#") ``` ## Primary sources All values pass through verbatim from these official sources. No transformations or imputations are applied. - [**FRED**](https://fred.stlouisfed.org/) — Federal Reserve Economic Data (St. Louis Fed) - [**Freddie Mac PMMS**](https://www.freddiemac.com/pmms) — Primary Mortgage Market Survey - [**BLS**](https://www.bls.gov/) — Bureau of Labor Statistics (CPI, employment, wages) - [**BEA**](https://www.bea.gov/) — Bureau of Economic Analysis (PCE) - [**US Treasury**](https://home.treasury.gov/) — Treasury yields - [**FDIC**](https://www.fdic.gov/resources/bankers/national-rates/) — National Rates and Rate Caps - [**EIA**](https://www.eia.gov/) — Energy Information Administration (oil, gas) - [**Federal Reserve H.10**](https://www.federalreserve.gov/releases/h10/) — Foreign exchange rates - [**CoinGecko**](https://www.coingecko.com/) — Cryptocurrency spot prices - [**World Bank Open Data**](https://data.worldbank.org/) — GDP, world inflation, world unemployment - [**IMF Commodity Data**](https://www.imf.org/external/np/res/commod/index.aspx) — Copper, corn (via FRED) ## Refresh cadence A GitHub Actions workflow ([`.github/workflows/refresh.yml`](.github/workflows/refresh.yml)) runs daily at 06:00 UTC, pulls the latest history from CalcFi's read-only API, and commits any changes. Most series update at their natural cadence: - Daily: Treasury yields, FX, oil, crypto, fed funds - Weekly: 30Y / 15Y mortgage, gas price - Monthly: CPI, PCE, unemployment, hourly earnings, commodities - Quarterly: credit card APR, personal loan - Annual: World Bank series ## License This repository's structure, scripts, and README are released under [CC0 1.0](https://creativecommons.org/publicdomain/zero/1.0/). The data itself in `datasets/*/data.csv` is licensed [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) — attribution requested. **How to attribute:** > Data via [CalcFi](https://calcfi.app), sourced from [primary source]. CC BY 4.0. Primary source terms also apply where data is mirrored from FRED, BLS, World Bank, etc. — consult the per-dataset `README.md` for the canonical source link. ## Why this exists Most personal-finance and macro data APIs are paid, rate-limited behind a credit card, or wrap a single primary source you could hit yourself. [CalcFi](https://calcfi.app) is a free personal-finance calculator platform that consumes these primary sources for its tools; this repo exposes the same data as a clean, version-controlled, Frictionless-compliant dataset bundle so it can be mirrored to DataHub, Hugging Face, Kaggle, and Zenodo. ## Mirroring this to other registries This repo is structured so each subdir under `datasets/` is a complete Frictionless Data Package. To publish on DataHub.io, point at this repo and set "Root Dir" to e.g. `datasets/30-year-fixed`. Repeat per dataset. For Hugging Face Datasets, Kaggle, Zenodo, or re3data — see [`docs/mirroring.md`](docs/mirroring.md) (TBD; PRs welcome). ## Contributing This is a mirror, not the source of truth. To propose data changes, request a new series, or correct a primary-source citation, open an issue or PR. For live calculator + API issues, see https://calcfi.app/about. ## Citation > Salmisto, J. (2026). CalcFi Open Data: 34 Free CC-BY Financial and Macro Time Series Mirrored from Primary Sources [Dataset]. Figshare. https://doi.org/10.6084/m9.figshare.32332290