Alpha and signal research
The best of a hundred backtests, part 2 of 3
AppliedMultiple choice
The best of a hundred backtests · Part 2 of 3
A researcher tries 100 variants of a signal on five years of daily data. In truth every variant has a Sharpe ratio of zero. The best variant shows an annualised in-sample Sharpe ratio of 1.3.
Roughly what best in-sample Sharpe ratio would 100 worthless, independent variants be expected to produce?
- AAbout zero, since each variant’s true Sharpe is zero
- BAbout 0.45, one standard error
- CAbout 1.1
- DAbout 1.35, from √(2 ln 100) ≈ 3.0 standard errors
The worked solution is in Premium
The answer, the full working and the one idea to take away – for this and all 1,322 questions in the bank. Answer it in practice and your working is marked, with a known mistake named when you make one.
Report an issue with this question
More alpha and signal research questions
- You backtest 20 independent strategies with no true skill over one year, where…Applied · Premium
- A researcher checks each new idea against the same held-out period and keeps…Applied · Premium
- A colleague shows you a backtest with a Sharpe ratio of 2.1 over three years.Advanced · Premium
- What does the probability of backtest overfitting measure?Expert · Premium
- A signal has an information coefficient of 0.05 and you make 400 independent bets a year.Applied · Free solution
- Your information coefficient is fixed and you want to double your information ratio.Applied · Free solution