WEBN & ACWI 0.61 correlation?
How come WEBN and ACWI calculated correlation is so low?
Looks pretty much the same to me!
https://portfolioslab.com/tools/stock-comparison/WEBN.DE/ACWI
1 ΠΊΠΎΠΌΠΌΠ΅Π½ΡΠ°ΡΠΈΠΉ
1 ΠΎΡΠ²Π΅Ρ
Π‘ΠΎΡΡΠΈΡΠΎΠ²Π°ΡΡ ΠΏΠΎ
Yeah, this pair is tricky. If you zoom in a bit, on the daily scale the prices start visibly fluctuate:
I think this is because the ETFs are traded in different markets with a different schedule. When ACWI is actively trading, the session for WEBN.DE is already over. So daily close prices will differ. That could lead to enough divergence on the daily price level to decrease the correlation value.
For pairs trading on different continents, a correlation calculated based on weekly data likely will be more accurate. We'll look if we can handle that automatically. Thanks for highlighting that!
Π² ΠΏΡ. ΠΌΠ΅Ρ.
Thanks, makes sense, I suppose combining the score for different time resolutions could be a way? I.e. assets moving closely together both on daily and weekly ranges should be considered more correlated than just one of the two?
Personally for the precedence of daily vs weekly I care more about weekly correlation. But that is a personal preference