Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have 10 years of backtested simulated performance of some trading strategy (using historical prices), and N months of actual trading performance. What statistical test can I do find out if I'm on target with the backtesting numbers? (both, in terms of expected annual returns and expected annual Sharpe ratio)

share|improve this question
up vote 6 down vote accepted

Letting $\psi_1, \psi_2$ be the sample Sharpe ratios of the two periods, the difference $\Delta \psi = \psi_1 - \psi_2$ is asymptotically normal. Under the null hypothesis that the population Sharpe ratios in the two periods are equal, the difference is asymptotically mean zero. The standard deviation is approximately $\sqrt{\frac{1}{120} + \frac{1}{n}}$, when your Sharpe ratios are 'annualized' to monthly terms. So the simplest test would be to reject the null if $|\Delta\psi|> 1.96 \sqrt{\frac{1}{120} + \frac{1}{n}}$.

My answer here is just a realization of @drnexus' answer to this question.

share|improve this answer
That's great. Thanks! – Nestor Jul 1 '11 at 17:04
I would like to amend my answer to state that a slightly more 'precise' version of the test is available via the upsilon distribution: . However, the 2-term approximation to the CDF of the upsilon is exactly the asymptotically normal approach given above. – shabbychef May 11 '15 at 16:57

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.