XLF historical data: 1-minute to daily bars, 2000 to 2026

XLF is the Financial Select Sector SPDR Fund; it tracks the financials sector of the S&P 500. MarketParquet holds XLF as a US ETF with 1-minute, 5-minute, 30-minute, 1-hour and daily OHLCV bars from 2000-01-03 to 2026-09-10 (6,713 daily bars), split-adjusted (dividends not embedded), as one Parquet file per trading day.

XLF is the most liquid sector ETF and the usual building block for sector-rotation backtests.

XLF at a glance

Sample: XLF 30-minute bars on 2026-09-16

The first regular-session bars from the newest 30-minute file, read live from the archive. Prices are the values as published; the full session has many more rows.

Open the full 2026-09-16 preview for XLF

Corporate actions in the changelog

No split adjustment for XLF appears in the published changelog, which lists every repair applied since it began. Earlier splits are already reflected in the archive: XLF prices are split-adjusted (dividends not embedded), so bars before any historical split sit on today's share basis. Dividends are deliberately not embedded; see data quality.

Load XLF in Python

Files are partitioned by trading day, so pulling one symbol means filtering across files. DuckDB reads the whole daily history in one statement; union_by_name and the COALESCE cover the older daily files, which carry a timestamp column where newer ones carry date:

  import duckdb

  xlf = duckdb.sql("""
      SELECT COALESCE(date, CAST(timestamp AS DATE)) AS date,
             open, high, low, close, volume
      FROM read_parquet('by_date/etf_daily/*/*.parquet', union_by_name=true)
      WHERE symbol = 'XLF'
      ORDER BY 1
  """).df()

Intraday files all share one schema, so Polars can scan a year of 5-minute bars lazily and keep only XLF:

  import polars as pl

  bars = (
      pl.scan_parquet("by_date/etf_5min/2025/*.parquet")
        .filter(pl.col("symbol") == "XLF")
        .select(["timestamp", "open", "high", "low", "close", "volume"])
        .sort("timestamp")
        .collect()
  )

Timestamps are US/Eastern bar open times. The Polars guide, the DuckDB guide and the backtrader guide take it from here: multi-year loads, resampling and feeding a backtest engine.

Available ETF files

Every file holds all ETF symbols for that day, XLF included.

Get XLF data

Free: a free account can download roughly the last year of daily ETF files, which include XLF, from the daily browse page or by API key. Complete Archive: every timeframe and the full history back to 2000-01-03 for XLF and every other symbol, $79 one-time.

FAQ

Is XLF history split-adjusted?

Yes. Prices are split-adjusted so the series is continuous across splits; dividends are not embedded, which keeps the bars equal to the prices that actually traded.

How far back does XLF go?

The first XLF bar in the archive is dated 2000-01-03 and the newest is 2026-09-10. New sessions are added the same night, typically by midnight ET.

What timezone are the intraday bars in?

US/Eastern. Regular hours are 9:30 AM to 4:00 PM ET; 1-minute files also carry pre-market from 4:00 AM and after-hours to 8:00 PM where trades occurred.

Can I download only XLF?

Files are per day, not per symbol, so you download the daily or intraday files for the dates you need and filter on symbol == "XLF". A year of daily files is small; the snippets above do the filtering in one pass.

Related

ETF data hub · browse ETF daily · all symbol pages · pricing

Other ETF pages: SPY · QQQ · IWM · DIA · TLT · GLD · SLV · ARKK · TQQQ · SQQQ · SOXL · UVXY · VXX