Wrong annualization of Sortino Ratio?

[gabrieleiacono]

The Sortino Ratio is annualized in stats.py as follows:
return sortino * ann_factor

where ann_factor is
ann_factor = annualization_factor(period, annualization)

I believe you actually meant to return
return sortino * np.sqrt(ann_factor)

just like you did with the Sharpe Ratio.

[twiecki]

Good catch, want to do a PR?

[rsheftel]

The math is actually correct in the existing version. The confusion is because of where the ann_factor multiplication occurs. The reason it is correct is because the calculation of the downside risk with the downside_risk function returns an annualized value, it does the * np.sqrt() in that function. The calculation of the numerator of returns in the sortino_ratio() function does not annualize:

mu = nanmean(adj_returns, axis=0) so the code annualizes at the end at the return value.
return sortino * ann_factor

Agree this is confusing, so I submitted a pull request that just moves the * ann_factor to the line that calculates the returns so the math is more obvious:

mu = nanmean(adj_returns, axis=0) * ann_factor

[twiecki]

@rsheftel Thanks, very glad you point this out before we merged it.

[twiecki]

Closed by #74.

[gabrieleiacono]

I'm terribly sorry, thank you @riabaldevia for finding my mistake. It is much better now, after PR #74.