Data quality

Validation Report

Generated 2026-08-20. Every check runs automatically on dataset creation. No manual edits, no cherry-picking.

35 Passed
2 Warnings
0 Failures

Dataset summary

Date range2007-01-03 → 2026-08-19
Trading days4,938
Years covered19.6
Tickers tracked (PIT)273
Tickers with OHLCV211
Tickers no OHLCV (acquired/delisted)64
Component change events228
Current NDX members103
Survivorship bias216 bps/yr
Survivor-only CAGR (2010–2026)19.36%
PIT-correct CAGR (2010–2026)17.2%

OHLCV price data

✓ Date range: 2007-01-03 → 2026-08-19 (4,938 trading days)
✓ Trading day count looks complete (4,938 ≥ 4,700)
✓ Tickers in file: 211
✓ Tickers with price data: 211
✓ All tickers have price data
✓ All 5 OHLCV fields present: close, high, low, open, volume
✓ Data is current — last bar 2026-08-19 (1 calendar days ago)
✓ No future-dated bars (no look-ahead contamination)
✓ Yearly coverage computed for 20 years
✓ No negative close prices
✓ No zero close prices

Point-in-time membership table

✓ PIT table shape: 4,938 days × 273 tickers
✓ NDX member count — min: 0, max: 107, current (2026-08-19): 103
✓ Current member count 103 is within expected range (95–105)
! 20 trading days with <90 NDX members (early-history membership gaps — see Known Limitations)
✓ No days with >110 NDX members
✓ AAPL is correctly marked as current NDX member
✓ MSFT is correctly marked as current NDX member
✓ AMZN is correctly marked as current NDX member
✓ GOOGL is correctly marked as current NDX member
✓ META is correctly marked as current NDX member
✓ RIMM is correctly marked as NOT in current NDX
✓ YHOO is correctly marked as NOT in current NDX
✓ ATVI is correctly marked as NOT in current NDX
✓ CELG is correctly marked as NOT in current NDX
! 64 tickers appear in PIT membership but have no OHLCV (acquired/delisted, documented in MISSING_TICKERS.txt)
✓ No calendar gaps > 5 days in trading day index

Component change history

✓ Component changes: 228 events (2007-02-01 → 2026-08-04)
✓ 'added_ticker' column present with 212 non-null entries
✓ 'removed_ticker' column present with 210 non-null entries
✓ History starts 2007-02-01 — covers pre-2008-crash period
✓ 24 changes in the last 18 months — data is being maintained

Included files

✓ ndx_ohlcv.parquet (31.5 MB) — OHLCV price data (30 MB)
✓ ndx_pit_daily.parquet (0.2 MB) — PIT membership table (binary, fast load)
✓ ndx_pit_summary.csv (0.0 MB) — Per-ticker membership dates
✓ ndx_component_changes.csv (0.0 MB) — Raw index add/remove events
✓ MISSING_TICKERS.txt (0.0 MB) — Documented unavailable tickers

Seen enough? This dataset passes every automated check — and documents every limitation.

The dataset is compiled from public records and sold as-is — this page is the complete list of what we know about its quality. If something matters to your research, verify it independently. Full terms.