Capital Markets Domain¶

Public equities market data with real S&P 500 tickers, daily OHLCV pricing via Geometric Brownian Motion, corporate actions, earnings, and insider transactions.

Tables¶

Table	Rows (small)	Description
`exchange`	3	Stock exchanges (NYSE, NASDAQ, AMEX)
`sector`	11	GICS sectors
`industry`	61	GICS industries with sector FK
`company`	100	Public companies with real S&P 500 tickers and SEC EDGAR CIKs
`daily_price`	~75,600	Daily OHLCV bars (252 trading days/year x 3 years x 100 companies)
`trade`	~300,000	Tick-level trades for streaming demos
`dividend`	~450	Dividend payments with ex-date, pay-date, frequency
`split`	~15	Stock splits with ratio (2:1, 3:1, etc.)
`earnings`	~1,200	Quarterly earnings with EPS estimate, actual, and surprise %
`insider_transaction`	~600	SEC Form 4 insider trades (BUY/SELL/GRANT)

Quick Start¶

from sqllocks_spindle import Spindle, CapitalMarketsDomain

result = Spindle().generate(domain=CapitalMarketsDomain(), scale="small", seed=42)
print(result.summary())

Key Features¶

Real tickers — 110 S&P 500 companies with actual CIK numbers, GICS sectors/industries, and exchanges
GBM pricing — daily OHLCV via Geometric Brownian Motion with enforced OHLC business rules (high >= low, close within range)
Log-normal volumes — realistic trading volume distributions (mean ~33K shares/day)
Zipf ticker distribution — tick-level trades concentrated in mega-cap names (alpha=1.5)
Dividend frequency — 75% quarterly, 20% annual, 5% special
Earnings surprise — EPS actual correlated to estimate (85-120%), surprise % normal(0%, 2.5%)
Insider titles — weighted distribution (CEO 15%, CFO 12%, Director 25%, VP 15%, etc.)
Star schema map — 4 dimensions (company, exchange, sector, industry) + 4 facts (daily_price, dividend, earnings, insider_txn)

Scale Presets¶

Preset	`company`	Years	Approx `daily_price` Rows
`fabric_demo`	30	1	~7,500
`small`	100	3	~75,600
`medium`	500	5	~630,000
`large`	1,000	10	~2,520,000
`xlarge`	4,000	20	~20,160,000