Datasets

COT positioning · Free+

Commitments of Traders

CFTC Commitments of Traders reports, all four report types. Legacy report since 1986; Disaggregated and Traders in Financial Futures since 2006. Published weekly Friday 3:30 pm ET with positions as of Tuesday. 431 markets.

Prefer a one-off bulk file? The full COT history (all four report types, futures-only and combined, 1986 to date) is also a free Parquet dataset on Hugging Face.

import finzdata as yf

cot = yf.Client().cot(report="legacy_fo", market="WHEAT")
print(cot.tail())

Browse live COT data

SEC fundamentals · Free+

Fundamentals from SEC XBRL

Company facts as reported to the SEC, parsed from XBRL filings. 10,435 SEC-reporting companies. Point-in-time queries via the as_of parameter return only what was public on that date.

import finzdata as yf

f = yf.Client().fundamentals(ticker="AAPL", concept="Revenues", as_of="2023-06-30")
print(f[["end", "value", "form", "filed"]])

Browse live fundamentals

Macro · Free · point-in-time on look-back

US macro series, with every vintage

158 US economic series from government sources — inflation, jobs, growth, rates, money and credit, housing, liquidity and financial conditions — refreshed the day each release comes out. Every revision is kept, so a backtest can use the number as it was first published instead of today's revised figure.

import finzdata as yf

c = yf.Client()
cpi_today = c.macro("CPIAUCSL")                        # current values
cpi_then = c.macro("CPIAUCSL", as_of="2022-06-30")     # what was known that day
upcoming = c.macro_calendar()                          # next release dates

FinzData Quantitative Model · Look-back plan

Macro regime labels, a nowcast and a market-stress probability

Our own model outputs, built only from public data. Every label follows a written rule applied to each input as it was published on the date, so the history has no hindsight in it. The nowcast estimates the period that has not been published yet; the published label is kept beside it so you can see both.

import finzdata as yf

c = yf.Client()
now = c.model("latest")                                  # today's labels, nowcast and stress probability
hist = c.model("labels", series="macro_quadrant")         # point-in-time history
rules = c.model("catalog")                               # the written rule behind every label

Model outputs are research data, not investment advice.

Licensed third-party series · Look-back plan

Credit, sentiment, housing and volatility series, with their sources

Series owned by data companies, served under licence from each owner on paid plans. Each download and each chart carries the owner's source credit.

import finzdata as yf

c = yf.Client()
c.licensed_catalog()          # every series, its owner, source line and availability
hy = c.licensed("BAMLH0A0HYM2")
curve = c.vix_futures(start="2024-01-01")

Money, liquidity & capital flows · Look-back plan

M1, M2, M3, central-bank liquidity and cross-border capital flows

The money and flow data macro traders watch, in one place and in consistent units (billions). Fed net liquidity, G3 central-bank balance sheets in dollars, broad money for the US, euro area, Japan and the UK, and Treasury TIC flows by country, ready to line up against prices in a backtest. Updated twice a day.

import finzdata as yf

c = yf.Client()
c.liquidity_catalog()                                   # every series, units, coverage, sources
liq = c.liquidity("FED_NET_LIQUIDITY", start="2015-01-01")
money = c.liquidity(["US_M2", "US_M3_ESTIMATE", "G4_BROAD_MONEY_USD"], transform="yoy", wide=True)
china = c.tic(country="China, Mainland", asset="treasuries")   # holdings and net purchases, USD bn

13F institutional holdings · look-back bundles

13F institutional holdings

Every Form 13F filed with the SEC since mid-2013: about 9,000 managers a quarter, every reported position, with CUSIP-to-ticker mapping and quarter-on-quarter flows. Amendments are applied the way the SEC intends (restatements replace the filing, new-holdings amendments add to it), and values are in dollars throughout.

import finzdata as yf
c = yf.Client()
c.institutional_ownership("NVDA")      # holders and flows, quarter by quarter
c.holders("NVDA", period="2026Q2")     # every manager holding it
c.manager_holdings(cik="1067983")      # one manager's whole book
c.managers(q="berkshire")              # find a manager

Browse all 13F managers

Daily prices · Free+

Daily US equity bars

End-of-day OHLCV bars. Stored unadjusted; adjustments are applied on read with adjust=none|splits|all. Delisted names are included on the look-back plan.

import finzdata as yf

df = yf.Ticker("MSFT").history(start="2000-01-01", auto_adjust=True)
print(df.tail())

Browse live prices

Intraday prices · hourly on every bundle · 1-minute and 5-minute from 25 years

Intraday bars: 1-minute, 5-minute, hourly

Intraday OHLCV bars on the ET wall clock, including pre-market and post-market sessions, unadjusted or adjusted for splits and dividends. Download free samples and read the full file specification before you buy.

import finzdata as yf

df = yf.Ticker("NVDA").history(interval="5m", start="2024-01-02")   # 1m, 5m, 1h
print(df.head())

Browse live prices

Dealer gamma · coming soon

GEX by ticker

Dealer gamma exposure computed from end-of-day options chains, aggregated per ticker.

import finzdata as yf

# Dealer gamma (GEX) is coming soon.
# Email [email protected] to join the early-access list.

Corporate actions · Free+

Splits and dividends

Splits and dividends delivered as dated events, so you can apply adjustments yourself or let the API do it on read.

import finzdata as yf

acts = yf.Client().actions("AAPL")
print(acts)

Browse live actions