Week 5: Understand Market Data
The quality and structure of market data can materially affect backtesting results. A backtest is only as dependable as the data and assumptions used to create it.
Learn about
- Open, high, low, close prices and trading volume
- Candle intervals and when a candle is actually complete
- Adjusted vs unadjusted stock prices and corporate actions
- Missing candles
- Futures contract rollover
- Options strike and expiry data
- Timestamp and time-zone handling
- Live data vs delayed data
Critical rule: an algorithm working with five-minute candles must know when a candle is complete. Using its final closing value before the interval has ended creates look-ahead bias — one of the most common backtesting errors.
Practical exercise
Download historical candle data for one liquid instrument and check:
- Whether any trading days are missing
- Whether timestamps are in IST
- Whether holidays are excluded
- Whether volume appears reasonable
- How stock splits and dividends are handled
- Whether the data covers the market conditions you need to test
Week 5 outcome
You should understand that clean, complete, correctly-timestamped data is a precondition for every result that follows.
This lesson is for educational purposes only. It is not investment advice, a recommendation or an assurance of returns. Trading and derivatives involve substantial risk, and backtested or paper-trading results do not guarantee future performance.
