Historical US Stock, ETF & Mutual Fund Prices in Snowflake

​

Overview

• 50,000+ US securities: ~18,500 stocks (common and preferred), ~5,950 ETFs and ~25,800 mutual funds
• Full history, back to each ticker’s listing date, often decades
• Raw and adjusted OHLC: as-traded open, high, low and close, plus versions adjusted for splits and dividends/distributions
• Refreshed daily at 5:00 AM Pacific, with a full per-ticker restatement twice a week
• One format for every asset class, so stocks, ETFs and mutual funds can be analyzed with the same query
• Automated pipeline quality checks (unique grain, placeholder price detection, corporate-action restatement)

​

Coverage

​

Schema

​

Security Reference

select code, exchange_code, name, exchange, type, asset_class
from security_vw
where code in ('AAPL', 'VTI', 'VFIAX')
  and exchange_code = 'US';

​

Daily Price History

select code, date, adjusted_open, adjusted_high, adjusted_low, adjusted_close, close, volume
from security_price_history_vw
where code = 'AAPL'
  and exchange_code = 'US'
order by date desc
limit 3;

​

Example: total return between two dates

with px as (
    select code, date, adjusted_close
    from security_price_history_vw
    where code in ('AAPL', 'VTI', 'VFIAX')
      and exchange_code = 'US'
      and date between '2016-01-01' and '2025-12-31'
)
select
    code,
    min(date) as start_date,
    max(date) as end_date,
    max_by(adjusted_close, date) / min_by(adjusted_close, date) - 1 as total_return
from px
group by code
order by total_return desc;

​

Example: latest close for every ETF

select s.code, s.name, p.date, p.close
from security_vw s
join security_price_history_vw p
  on p.code = s.code
 and p.exchange_code = s.exchange_code
where s.asset_class = 'etf'
  and not s.is_inactive
qualify row_number() over (partition by s.code, s.exchange_code order by p.date desc) = 1;

​

Data Ingestion

​

Data Quality

• Unique grain: every (code, exchange_code, date) must be unique in the price history, and every (code, exchange_code) must be unique in the reference table. Both are checked automatically on every run.
• Placeholder price detection: zero adjusted closes and the source’s backward-adjustment overflow value are flagged in the pipeline and excluded from security_price_history_vw instead of being passed through as real prices.
• Stub row detection: a security’s very first price row is excluded when it sits more than 4 days before the next one, a lone placeholder well ahead of where real trading begins.
• Corporate-action restatement: new splits and dividends trigger a full history refetch for that security, and a twice-weekly full restatement catches anything the incremental path missed.

​

Disclosures

​

F.A.Q.