Hi Brian,
I just updated the daily us-stock bundle ran a backtest and got very strange returns from yesterday's data (8-07-2024). Is there something wrong with the updated prices from yesterday?
I don't believe there are any issues with yesterday's prices - not that I am aware of. I would dig into your returns and positions to pin down where the performance spike is coming from in more detail. If a price looks wrong based on that, please report it.
It looks like the split for Microstrategy has an issue !
Matched Sids: ['FIBBG000GQJPZ0']
Field Date
Close
2024-08-02 144.799
2024-08-05 130.900
2024-08-06 136.921
2024-08-07 1246.850
The ex date is a day early on this split. Resolution in progress...
This should now be resolved, with the corrected split pulled in with the 2024-08-08 daily update.
Brian,
I am kind of confused about the new prices. When I query the table now I get
Matched Sids: ['FIBBG000GQJPZ0']
Field Date
Close
2024-08-02 14.480
2024-08-05 13.090
2024-08-06 13.692
2024-08-07 124.685
2024-08-08 135.990
Name: FIBBG000GQJPZ0, dtype: float64
This was a 10-for-one split, so shouldn't the old prices be in the 100s range? The price was around 1300 pre-split so this is very confusing!
I'm getting the expected results:
In [1]: from quantrocket import get_prices
...: get_prices("usstock-1d-bundle", fields="Close", sids="FIBBG000GQJPZ0", start_date="2024-08-01")
Out[1]:
Sid FIBBG000GQJPZ0
Field Date
Close 2024-08-01 151.181
2024-08-02 144.799
2024-08-05 130.900
2024-08-06 136.921
2024-08-07 124.685
2024-08-08 135.990
Is there anything unusual about how you're querying? Your results imply that the erroneously dated split is still on your system, resulting in a double split of the pre-Aug 7 prices. But I don't know why it would be.
If this is the daily bundle, maybe try dropping and re-ingesting it.
Dropping the bundle fixed it!
Thanks
This topic was automatically closed after 4 days. New replies are no longer allowed.