Skip to content
QuantGPT
← All datasets

US stocks

US stock and fund prices

QuantGPT

Daily open, high, low, close and volume for US stocks, funds and the main indexes, raw and adjusted.

Use in my models: choose in My dataAll my data choices

What it is

Daily open, high, low, close and volume for US stocks, funds and the main indexes, raw and adjusted.

  • prices daily: Daily open, high, low, close and volume, raw and adjusted (splits and dividends), VWAP, dollar volume, shares and market cap: US stocks, ETFs and the indexes ($DJI from 1896, $SPX from 1928, $SPXTR from 1988).
  • prices daily, month ends: Adjusted close at each month end (the backtest grid).
  • prices daily, week ends: Adjusted close at each week end.

Source and licence

Source: EODHD.

Licence: licensed vendor: SHOW_VENDOR_PRICES / SHOW_VENDOR_DATA. This is our reading of the table; counsel decides.

Credit: EODHD end of day, via the QuantGPT engine; coins: Binance daily closes (Data sources).

The app shows it only while the server switch SHOW_VENDOR_PRICES is on.

Coverage and gap years

54,854 names, 97,335,854 rows in prices daily, from May 26, 1896 to Oct 8, 2026.

Years with no rows: none between the first and the last.

TableRowsFirstLast
prices daily97,335,854May 26, 1896Oct 8, 2026
prices daily, month ends3,156,203Jan 31, 1928Oct 8, 2026
prices daily, week ends13,628,671Jan 6, 1928Oct 8, 2026

Freshness

Last row Oct 8, 2026: 1 day behind on the day measured. Updates: every trading day.

Point in time

Each bar is keyed on its trading day, with the raw close beside the adjusted one.

Adjusted closes are rewritten after each split or dividend (the repairs are on Data status); raw closes stay as traded.

Delisted names

Companies that later delisted are in from 1998; before that the history is survivors only, so early years lean toward the winners.

Known issues

  • Early price history is survivors only. prices daily reaches 1896 for $DJI and 1928 for $SPX. Single stocks are thin before 1980. Delisted companies are in from 1998, so month-end tables are thin from 1929 to 1961. (WAREHOUSE, Gaps and freshness that matter)
  • Share errors reach market cap. prices daily shares out is the latest shares diluted, and market cap is raw close times that figure. JPM on 2026-10-08: 2,793.7M x 331.42 = 925.9B. With SEC's latest 2,694.2M it would be 892.9B (+3.7%). (ACCURACY, Quarter by quarter (2024-06 to 2026-07, 44 company quarters with an SEC figure))
  • Index rows have adj close equal to raw close (price indexes, no dividends). $SPXTR is the total return series. (ACCURACY, 2. Price facts)
  • Prices are stored at single precision: 4.40 reads 4.400000095367432. This is harmless at two decimals; round on display. (ACCURACY, 2. Price facts)
  • Dividend spot check (corporate actions): AAPL 0.26 then 0.27 from 2026-05-11. JPM 1.50 then 1.65 on 2026-10-06. MSFT 0.91 then 0.98 on 2026-11-19. XOM 1.03. The yield factor agrees: XOM div yield TTM 2.45% = 4 x 1.03 / 168.50. (ACCURACY, 2. Price facts)
  • P/E spot check: AAPL pe TTM 38.95 against 340.42 / (1.85 + 2.84 + 2.01 + 2.02) = 39.04. The factor is monthly, so it uses the month's formation price. Note it is built on the reported (adjusted) EPS above. (ACCURACY, 2. Price facts)
  • 48 splits are still unadjusted, 30 of them in 2026. (ACCURACY, 5. Broken splits)
  • 27 of the 48 have a readjust done row and still jump. (ACCURACY, 5. Broken splits)
  • 10 are flagged warrants, units, preferred or non-operating. (ACCURACY, 5. Broken splits)
  • 17 carry a market cap, and some caps are far too large because the share count did not follow the split. (ACCURACY, 5. Broken splits)

Where it is used

Engine tools that reach it (your AI, over MCP):

  • get_data Standard warehouse series without SQL.
  • run_backtest Backtest any criteria in seconds: up to 8 filters over the factor vocabulary, ranked by a numeric factor, top N equal weight (or weighting: market cap, inverse vol, factor, custom, with max weight),...
  • run_market_timing Classic index timing rules on a benchmark: sma10m (10-month SMA), goldencross, sellinmay, santa.
  • query_warehouse Run read-only SQL (DuckDB dialect) directly against any research database.

Ask AI

Every count and date here was measured on Oct 9, 2026 from the QuantGPT engine's warehouse and is read from the catalogue file when the site is built. Metadata only: no vendor rows are shown.