Historic Stock Options Dataset

This repository contains historical options data for three major U.S. equity ETFs: SPY (S&P 500), IWM (Russell 2000), and QQQ (Nasdaq-100). The dataset spans January 2008 to December 2025 and includes over 53 million option contracts with Greeks, implied volatilities, and market microstructure variables.

Overview

Symbol	Description	Records	Period	Compressed
SPY	SPDR S&P 500 ETF Trust	24.7M	2008-2025	1.5 GB
IWM	iShares Russell 2000 ETF	13.4M	2008-2025	812 MB
QQQ	Invesco QQQ Trust	15.3M	2011-2025	965 MB
Total		53.4M		3.3 GB

The dataset covers several market regimes including the 2008 financial crisis, the 2020 COVID-19 volatility spike, and the subsequent recovery periods.

Data Format

Data is provided in two formats. SQLite databases are compressed with Zstandard and require decompression before use (see DECOMPRESSION.md). Parquet files are partitioned by year and can be used directly.

SQLite

File	Compressed	Decompressed
data/spy_options.db.zst	1.5 GB	9 GB
data/iwm_options.db.zst	812 MB	4.8 GB
data/qqq_options.db.zst	965 MB	5.6 GB

Parquet

Directory	Size
data/parquet_spy/	559 MB
data/parquet_iwm/	313 MB
data/parquet_qqq/	638 MB

Schema

The options data table contains the following columns:

Column	Type	Description
contract_id	TEXT	Unique contract identifier
symbol	TEXT	Underlying symbol
expiration	TEXT	Expiration date
strike	REAL	Strike price
type	TEXT	Call or put
bid, ask	REAL	Best bid and ask prices
volume	INTEGER	Daily trading volume
open_interest	INTEGER	Open interest
date	TEXT	Observation date
implied_volatility	REAL	Black-Scholes implied volatility
delta, gamma, theta, vega, rho	REAL	Option Greeks

Each database also includes an underlying_prices table with daily OHLCV data.

Usage

Parquet

import pandas as pd

# Load single year
df = pd.read_parquet('data/parquet_spy/options_2024.parquet')

# Load all years
from glob import glob
files = sorted(glob('data/parquet_spy/options_*.parquet'))
df = pd.concat([pd.read_parquet(f) for f in files])

SQLite

import sqlite3
import pandas as pd

conn = sqlite3.connect('data/spy_options.db')
df = pd.read_sql("SELECT * FROM options_data WHERE date >= '2024-01-01'", conn)

Documentation

For detailed statistics and methodology, see the technical report in reports/dataset_description.pdf.

Citation

@misc{dubach2025options,
  author = {Dubach, Philipp},
  title = {Historic Options Dataset: SPY, IWM, and QQQ Options 2008-2025},
  year = {2025},
  publisher = {GitHub},
  url = {https://github.com/philippdubach/historic-options-dataset}
}

License

MIT License. See LICENSE for details.