US ETFs — Intraday and End-of-Day OHLCV

Schema

Data formats

• Intraday bars (1-minute, 5-minute, 30-minute, 1-hour): the timestamp has the date and time, yyyy-MM-dd HH:mm:ss (US Eastern).
• End-of-day bars (daily): the timestamp is the date only, yyyy-MM-dd, with no time.
• Columns: timestamp, open, high, low, close, volume (shares), the same at every frequency.

Intraday bars (1-minute example)

timestamp,open,high,low,close,volume
2026-10-02 04:00:00,766.8,767.27,766.04,767.08,4630
2026-10-02 04:01:00,767.03,767.61,766.01,766.2,5072
2026-10-02 04:02:00,766.28,767.65,765.43,767.33,22908

End-of-day bars (daily)

timestamp,open,high,low,close,volume
2026-10-02,770.58,772.65,767.145,769.64,46335295
2026-10-05,769.69,776.605,769.63,774.83,44840163

One CSV per ticker, frequency and adjustment type, delivered as bulk zip archives.

Methodology

Every ticker comes in three variants: unadjusted (1-minute and daily bars only), split-adjusted, and split + dividend-adjusted. For splits and reverse splits, all earlier prices are adjusted by the split ratio, and volumes are adjusted by the same ratio. For dividends, all prices before the ex-dividend date are multiplied by 1 − (dividend ÷ prior close). The prior close is the last traded price on the day before the ex-date, not the regular-session close. Volumes are adjusted too. Since August 2023, splits and dividends have been taken directly from SEC filings rather than third-party feeds. Bars cover the extended session, including pre- and post-market trades.

Intended uses

• Training and evaluating forecasting and time-series models
• Building reinforcement-learning environments and market simulators
• Backtesting and quantitative research

Known limitations

• Bars with zero volume are left out rather than filled, so a gap in the sequence means no trades. Gaps are most common in small caps and in pre- and post-market hours.
• Trading halts, such as the March 2020 market-wide circuit breakers, show up as gaps.
• Only ETFs that are currently listed are included. Closed ETFs are not, so a universe built from this data has survivorship bias.
• Unadjusted data is only available as 1-minute and daily bars.

How to cite

@misc{frd_us_etfs_intraday,
  title  = {US ETFs — Intraday and End-of-Day OHLCV},
  author = {FirstRate Data},
  year   = {2026},
  url    = {https://firstratedata.com/ai/datasets/us-etfs-intraday}
}