0
$\begingroup$

I'm currently creating a backtesting script and I've got to the point of calculating risk metrics.

It seems like the interval (daily, weekly, or monthly) I use for returns heavily changes the outputted Sharpe and Sortino ratios.


I have date for daily returns between 2018-08-31 and 2021-08-31.

The risk-free rate I've used is 1.36% annually.


Daily

Sharpe: 0.14
Sortino: 0.30

Weekly

Sharpe: 0.26
Sortino: 0.95

Monthly

Sharpe: 0.53
Sortino: 4.79


If this difference is to be expected, what's the gold-standard interval in the industry?

If this difference isn't expected, what errors may have occurred to cause it?

NB: The returns are extremely volatile and I'm guessing this may be the root cause.

$\endgroup$
3
  • 2
    $\begingroup$ Are you annualizing/rescaling your metrics? That would be the first thing to check, however, even if you do that, they will still be different. See : alo.mit.edu/wp-content/uploads/2017/06/… for a great discussion. $\endgroup$ Sep 17, 2021 at 21:51
  • $\begingroup$ @rubikscube09 - wow, I totally forgot about that! Is it best to annualize from daily? And does the same apply for Sortino? $\endgroup$
    – Alex Vale
    Sep 17, 2021 at 21:59
  • $\begingroup$ The annualized values of Sharpe will be $0.14 \sqrt{252},0.26 \sqrt{52},0.53 \sqrt{12}$ or 2.22, 1.87, 1.83 respectively. They are fairly consistent with each other. $\endgroup$
    – nbbo2
    Sep 18, 2021 at 7:59

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.