Dear @cortezkwan,
Users can use their own files, so yes, it is possible to use your pre-trained weights or a model by uploading it into strategy environment filesystem and later using it in your code. The only important thing is that the output weights should be in the correct format that matches our structure.
Regards
@kairos Hi,
thanks for pointing this out, we are working on it, it will be fixed as soon as possible.
The problem is, that for some dates prior to 2014 we don't get OHLC prices at all for most assets (all NaNs, so is_liquid is NaN too). We have already reported it to our data provider. Anyway, the clean() function should work properly.
Best regards
@omohyoid Hi,
To prevent potential forward-looking behavior in the strategy, we use day-by-day submission processing. This means that when defining weights for a specific date, only data up to that date is available. Please ensure that data used by your model is handled correctly within the strategy.
Any use of future data for a specific date will produce different results than expected, and the strategy will not be eligible for competition.
Additionally, using external data sources or data not provided by Quantiacs is not allowed.
Regards
Dear @darwinps,
The difference is that you call filter method in your code where you see this warning. When that method is called, it calculates stats internally at that point in time, so that's when you actually see the warning, since at that time, the liquidity filter hasn't been applied.
The final stat calculation that you are doing in your code (after the liquidity filter) is not giving any warnings which is expected behavior.
Best regards
Dear @omohyoid,
Currently we are facing some data issues, the error is not caused by changes in qnt library. At the moment, only current index members of S&P500 index are available, but Nasdaq100 stocks are temporarily unavailable. Sorry for the inconvenience, we are working on a solution.
Dear @darwinps,
This discrepancy is normal due to different ways how the stats are calculated at the beginning of the in sample period. Over a longer period, these statistics should become very similar or identical.
Thanks for bringing this up, and if you notice some big discrepancies in the future, please report that to us.
Best regards
@carogate Hi, at the moment unfortunately not, sorry...
@carogate Hi, please take a look at this example:
https://github.com/quantiacs/strategy-predict-NASDAQ100-use-SPX/blob/master/strategy.ipynb
@sun-73 Hi, there is no limit to the maximum limit to the number of stocks. The cap of 10% stays the same. Call documented in:
https://quantiacs.com/documentation/en/data/stocks.html#stocks-s-p500
@omohyoid Hi, we do not have such calls, sorry
Dear @omohyoid,
You can remove that warning by altering the code of the qnt library which is generally not advised since it can affect many things in your local or online development environment and could lead to different results than expected after submission. For your specific case, you can just remove the line for logging that warning in the qnt/stats.py file in the calc_stat function.
Regards
@Sun-73
Dear Sun-73,
Thank you for pointing this out. The submission seems to have been stuck, but it should be processed soon.
Best regards
@sun-73 Hi, sorry for the gap, we are finalizing the new contest and will be ready in a couple of weeks.
@sun-73 HI, it is ok, the queue is huge, as usual as a competition approaches the end. All submissions submitted before the deadline will be processed.
@omohyoid Dear omohyoid,
Yes, that's right. After submitting your strategy shouldn't override environment variables.
Regards
@theflyingdutchman Hi, the Q20 had very few participants, less than 30, and no strategies are being traded at the moment according to the contest rules. Indeed we made no official announcement.
However, the participants had the opportunity to submit the same code (provided it is correct) to the Q21 (in other words no correlations check are performed).
It is very cumbersome to remove strategies from the web page, so they are currently displayed.
If your strategy will continue to perform good and we believe it can be traded, we will contact you.
@commanderangle Dear commanderangle,
You can now find your strategies in the "In contest" tab since the contest has begun.
Regards
@eddiee Dear Eddiee, the issue is not the low amount of assets, although it is better to trade more assets, but the static vs. dynamic selection.
Please note that the website shows the performance of the strategies independently on the fact that are still currently traded or not, and it uses the original volatility (in other words it does not reflect any scaling of the volatility).
@omohyoid Dear omohyoid,
Your strategy should not override environment variables after submitting as it can lead to unexpected behaviour like this.
Regards
@omohyoid Dear omohyoid,
There was some downtime yesterday, sorry, this issue is fixed. Thank you for your patience.
Regards