From trading ideas to honest backtests
4 readings
Choose a reproducible strategy, specify the trading rule, and test what remains after costs and estimation uncertainty.
- Who this is for
- Readers who can use Python and want to turn a trading paper into a research project.
- What you will work towards
- Write a testable strategy specification and evaluate an out-of-sample backtest with realistic costs.
Before you start
Use these SMQ lessons to check the ideas the readings assume. Lesson pages show the outline and how to access the full lesson.
- Expectation, Mean, and Variance
Read expected payoffs and risk estimates in the same mathematical language.
- Realistic Backtesting: Transaction Costs, Borrow, and Survivorship Bias
Check whether a paper’s signal survives the costs and information available at the time.
Read in this order
Quantitative Trading: How to Build Your Own Algorithmic Trading Business
Start with the practical research process and data decisions.
151 Trading Strategies
Choose one explicit strategy rather than attempting to implement the whole survey.
Time series momentum
Study a specific empirical strategy after surveying the possibilities.
The Statistics of Sharpe Ratios
Evaluate how much uncertainty sits behind a reported performance number.
Continue your study on SMQ
- Realistic Backtesting: Transaction Costs, Borrow, and Survivorship Bias
Check whether a paper’s signal survives the costs and information available at the time.
- Backtest Statistics: Sharpe Ratio, Probabilistic SR, and Deflated SR
Separate an estimated Sharpe ratio from evidence that a strategy has skill.
Follow the connected lessons, worked examples, and exercises in the app. Full lesson access depends on the topic and your membership.
Explore statistical arbitrage