Backtesting
Quantitative Finance glossary · Quantitative Finance
Definition
Out-of-sample evaluation of a trading rule's historical performance — strictly distinct from in-sample fitting. The four backtest killers: (i) look-ahead bias (using data not available at decision time, e.g. restated fundamentals or close prices stamped at the open), (ii) survivorship bias (delisted firms missing from the universe), (iii) data-mining bias / multiple-hypothesis testing (the deflated Sharpe ratio DSR corrects for the number of trials), and (iv) regime-dependence. The Bailey-López de Prado 'minimum backtest length' formula gives the number of independent trials needed before a reported Sharpe is statistically distinguishable from luck.
Backtesting in plain English
Backtesting is testing a trading idea on historical data to see if it *would* have worked — but done honestly, meaning you only let the strategy 'see' information it could have actually known at the time. Four traps quietly turn a fake winner into a real loser: peeking at data from the future, ignoring companies that went bust and disappeared from your list, and — the sneakiest — trying hundreds of strategies and cheering for whichever got lucky. That last one is like flipping thousands of coins and bragging about the one that landed heads ten times; with enough tries, dumb luck *guarantees* an impressive-looking result. So statisticians built formulas that adjust for how many ideas you secretly tried, telling you how good a result must be before it's more than coincidence.
Learn these first
Related Quantitative Finance terms
Quantitative Finance practice questions · Quantitative Finance flashcards · Quantitative Finance study guide