Yahoo Finance Integration¶

Free historical market data from Yahoo Finance.

Overview¶

Yahoo Finance provides:

Free historical prices (OHLCV)
Dividends and splits
Basic fundamentals
Wide symbol coverage

Basic Usage¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT", "GOOGL", "AMZN", "META"]
  start = "2020-01-01"
  end = "2024-12-31"

Configuration¶

Symbols List¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT", "GOOGL"]
  start = "2020-01-01"

From File¶

Text Only

data:
  source = "yahoo"
  symbols_file = "universe.txt"  # One symbol per line
  start = "2020-01-01"

Index Components¶

Text Only

data:
  source = "yahoo"
  index = "^GSPC"  # S&P 500 components
  start = "2020-01-01"

Data Fields¶

Available Fields¶

Field	Description
`date`	Trading date
`open`	Opening price
`high`	Daily high
`low`	Daily low
`close`	Closing price
`adj_close`	Adjusted close
`volume`	Trading volume

Default Mapping¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT"]
  start = "2020-01-01"
  columns:
    adj_close: Numeric as prices  # Use adjusted close
    volume: Numeric

OHLCV Data¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL"]
  start = "2020-01-01"
  columns:
    date: Date
    symbol: Symbol as ticker
    open: Numeric
    high: Numeric
    low: Numeric
    adj_close: Numeric as close
    volume: Numeric

Adjusted Prices¶

Yahoo provides split and dividend adjusted prices:

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL"]
  columns:
    adj_close: Numeric as prices  # Adjusted for splits/dividends
    close: Numeric as raw_prices  # Unadjusted

Always use adj_close for backtesting.

Corporate Actions¶

Dividends¶

Text Only

data dividends:
  source = "yahoo"
  type = "dividends"
  symbols = ["AAPL", "MSFT"]
  start = "2020-01-01"

Splits¶

Text Only

data splits:
  source = "yahoo"
  type = "splits"
  symbols = ["AAPL", "MSFT"]
  start = "2020-01-01"

Caching¶

Enable Caching¶

YAML

data:
  yahoo:
    cache:
      enabled: true
      directory: /var/cache/sigc/yahoo
      ttl_hours: 24

Clear Cache¶

Bash

sigc cache clear --source yahoo

Rate Limiting¶

Default Limits¶

Yahoo has rate limits:

~2000 requests per hour
~100 requests per minute

Configure Rate Limiting¶

YAML

data:
  yahoo:
    rate_limit:
      requests_per_minute: 30
      retry_after_seconds: 60

Error Handling¶

Common Issues¶

Error	Cause	Solution
`Invalid symbol`	Symbol not found	Check ticker spelling
`No data`	Symbol delisted or no history	Use alternative symbol
`Rate limit`	Too many requests	Enable caching, reduce frequency

Fallback Data¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT"]
  options:
    on_error = "skip"  # skip | fail
    fallback = "cache"

Large Universes¶

Batch Downloading¶

YAML

data:
  yahoo:
    batch_size: 50       # Download in batches
    parallel: true        # Parallel requests
    max_workers: 4

Progress Tracking¶

Bash

sigc data fetch --source yahoo --symbols sp500.txt --progress

Example Strategies¶

Basic Momentum¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT", "GOOGL", "AMZN", "META",
             "NVDA", "TSLA", "JPM", "JNJ", "V"]
  start = "2020-01-01"
  columns:
    adj_close: Numeric as prices
    volume: Numeric

signal momentum:
  emit zscore(ret(prices, 60))

portfolio main:
  weights = rank(momentum).long_short(top=0.3, bottom=0.3)
  backtest from 2020-01-01 to 2024-12-31

With Volume Filter¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT", "GOOGL"]
  start = "2020-01-01"
  columns:
    adj_close: Numeric as prices
    volume: Numeric

signal momentum:
  // Filter low volume stocks
  avg_vol = rolling_mean(volume, 20)
  liquid = avg_vol > 1000000

  mom = zscore(ret(prices, 60))
  filtered = where(liquid, mom, 0)

  emit filtered

Using Technical Indicators¶

Text Only

data:
  source = "yahoo"
  symbols = ["AAPL", "MSFT", "GOOGL"]
  start = "2020-01-01"
  columns:
    adj_close: Numeric as prices
    high: Numeric
    low: Numeric

signal mean_reversion:
  // RSI signal
  rsi_val = rsi(prices, 14)
  rsi_sig = -zscore((rsi_val - 50) / 25)

  // ATR for position sizing
  atr_val = atr(high, low, prices, 14)
  size_adj = 1 / atr_val

  emit rsi_sig * zscore(size_adj)

Symbol Lookup¶

US Stocks¶

Use ticker symbol: AAPL, MSFT
Include exchange suffix for clarity: AAPL.US

International¶

UK: VOD.L (London)
Germany: BMW.DE (Frankfurt)
Japan: 7203.T (Tokyo)

ETFs¶

SPY, QQQ, IWM, VTI
International: EFA, EEM, VEA

Indices¶

S&P 500: ^GSPC
Dow Jones: ^DJI
Nasdaq: ^IXIC
VIX: ^VIX

Limitations¶

Data Quality¶

May have occasional errors
Historical data can change
Corporate actions may be delayed

Coverage¶

US stocks: Excellent
International: Good
OTC/Pink sheets: Limited

Real-time¶

15-20 minute delay for quotes
Not suitable for live trading data

Best Practices¶

1. Use Adjusted Close¶

Text Only

columns:
  adj_close: Numeric as prices  # Always

2. Enable Caching¶

YAML

cache:
  enabled: true
  ttl_hours: 24

3. Verify Data Quality¶

Text Only

signal quality_check:
  // Check for suspicious data
  daily_ret = ret(prices, 1)
  suspicious = abs(daily_ret) > 0.5

  emit where(not(suspicious), prices, 0)

4. Handle Missing Symbols¶

YAML

options:
  on_error = "skip"

5. Use for Research, Not Production¶

For production, use a professional data provider.

Python Integration¶

Python

import pysigc
import yfinance as yf

# Fetch data with yfinance
tickers = ["AAPL", "MSFT", "GOOGL"]
data = yf.download(tickers, start="2020-01-01")

# Convert to sigc format
prices = data["Adj Close"].reset_index().melt(
    id_vars=["Date"],
    var_name="ticker",
    value_name="prices"
)

# Run strategy with custom data
results = pysigc.run("strategy.sig", data={"prices": prices})

Next Steps¶

CSV Format - Using local CSV files
Parquet Format - Faster data loading
Streaming - Real-time data