Country
You are in the Nigeria edition. Type your country to switch.
Nigeria
Loading every country…
No country matches. Check the spelling, or choose "Any country" for the Global edition.
Your learning profile
Algorithmic trading · Lesson 3 of 8
Data hygiene for backtests: point-in-time data, survivorship, look-ahead and corporate actions
A backtest is only as honest as its data. Prices that were adjusted later, companies that vanished from the list, figures that were published weeks after the date they describe: each one quietly flatters a strategy. This lesson shows how each bias creeps in and how to keep it out.
This lesson is complete in writing. A video for it is on its way.
What you will learn
- By the end you can explain point-in-time data and why revised figures flatter a backtest.
- You can spot look-ahead bias in a rule, including the trap of trading at a price used to make the signal.
- You can explain survivorship bias and why delisted companies must stay in the test.
- You can check that splits, dividends, ticker changes and other corporate actions are handled consistently.
Key terms in this lesson
- Backtest
- A test of how a strategy would have performed on past data.
- Point-in-time data
- Data that records each figure with the date it became public and keeps the original value even after restatements, so historical tests use only what was known at the time.
- Look-ahead bias
- Using information in a test that would not have been known at the time.
- Survivorship bias
- Distorted results from studying only the survivors, such as companies still listed, and ignoring those that failed.
- Corporate action
- An event decided by a company that affects its shares, such as a dividend, bonus, rights issue or split.
- Step 1: A backtest is only as honest as its data, not started
- Step 2: Point-in-time data and look-ahead bias, not started
- Step 3: Survivorship: testing only the winners, not started
- Step 4: Corporate actions, not started
- Step 5: A data checklist before any test, not started
About 10 min of reading
Step 1 of 5
A backtest is only as honest as its data
Opening this step…
Step 2 of 5
Point-in-time data and look-ahead bias
This step opens when you finish the one above.
Step 3 of 5
Survivorship: testing only the winners
This step opens when you finish the one above.
Step 4 of 5
Corporate actions
This step opens when you finish the one above.
Step 5 of 5
A data checklist before any test
This step opens when you finish the one above.
Key takeaways
The key takeaways open when you finish every step.
Lesson quiz
5 questions. You need 85% to pass.
The quiz opens when you finish every step.
Read first
Stockertim Learn is education. It explains how investing works and never recommends buying or selling anything.
Popular this week: Basics: BeginnerNGX & Africa: Beginner