Posts made by support

@nosaai Hi,

which version of qnt library is used? We introduced that function about a year ago, with S&P500 stocks dataset, maybe try with the most recent qnt version. If the issue persists, please let us know.

Regards

@alexeigor Hi,

sorry, it should work now.

Best regards

Hi,
thanks for pointing this out, we are working on refactoring the code, qnt is reverted to previous version. Sorry for late answer.

Dear @omohyoid,

That would not be a quantitative approach, hence it is not allowed based on the current contest rules.

Best regards

Dear @angusslq,

The ranking is calculated based on contest sharpe ratio only. The situation that you saw at that time was only temporary because at that point the calculations for new day were ongoing, so some strategies were processed before others, and their contest sharpe was updated, but the ranking is only updated after every single strategy from that contest has been processed and updated.

Regards

Dear @silverstar1003,

There was a temporary problem with our servers which has been resolved, and your strategy should be calculated soon.

Regards

Dear @angusslq,

Once the competition ends its live period (currently 4 months for Q22), the prizes are given. That means that at the end of those 4 months we sort all strategies and only the top 7 by sharpe ratio are eligible for prize and get allocation: 1st place 1M, second place 500k etc. and this cannot be changed afterwards. The prizes are not given on the daily basis and certainly not during the contest live period. You can find more info in the contest rules page on our website.

For your second question, we assume risk-free rate to be zero. You can find additional information about how we use sharpe ratio here.

Regards

Dear @captain-prairie_dog,

Yes, that's right. We will announce next contest soon, but roughly, the deadline period will be a couple of months away, so the users get enough time to develop their strategies.

Regards

@dark-pidgeot yes

Dear @dark-pidgeot,

There are two ways of how you can select your strategies to participate in the contest.
The first way is to use our auto-select feature. If it's turned on, then the top 15 strategies by performance will be automatically selected for participation.
The second way is to disable auto-select feature and use manual selection, after which you will have to hand-pick the strategies that you would like to participate.

It appears that you used manual selection at that time but didn't select any strategy manually. After careful review we concluded that you didn't break any other rule and your strategies are eligible for participation, and because this is a mistake users often make, we will add your submissions to the Q22 contest.

Regards

Dear @captain-prairie_dog,

Each contest has it's deadline period, for Q22 it was 31st of January. The live period is 4 months during which strategies will be evaluated and at the end of that period the prizes will be given to the winners. So to answer your question, yes, user needs to backtest and submit the strategy before the submission deadline and pass the filters (sharpe ratio, liquidity etc.) in order to participate in the contest. After that, all participating strategies are evaluated on real data in the following contest period (usually 4 months) and the score is calculated as a sharpe ratio. You can check what it is and how is it calculated in the documentation page on our website.
Hopefully this will clear up the confusion, and if you have any further questions, please feel free to ask.

Best regards

Dear @angusslq,

Allocations are set at the end of the contest, and they take into account the strategies ranking at that time. We update the rankings as we process the participating strategies with the new available data, which happens 5 times a week for 5 working days. The allocations that you refer are probably in the systems page and they take into account the current ranking in the contest but that doesn't mean anything until the contest ends and prizes are given.

Regards

Dear @svyable,

If you use our backtest function, the liquidity filter will applied automatically, you can just ignore that warning.
On the other hand, if you decide to use single-pass strategy option you should apply that filter manually or just use clean function that has that option included.

Regards

Dear @angusslq

Yes, there is a large queue currently as we are nearing the end of the submission period for the following contest. All submissions submitted before the deadline will be eligible for participating in the contest, including yours.

Regards

@auxiliary-snail Hi,

unfortunately, this is not allowed and in accordance with the rules. Using hard-coded time periods in which trading algorithm will work differently, is not a quantitative method (just like manual asset selection, e.g. "trade only Apple or Microsoft"). We still haven't implemented a mechanism for automatic recognition of such behaviors in trading strategies, and even though a strategy could be successfully submitted, it will not be eligible for prize winning.
What we are searching for, is well performing strategy over entire in_sample period (SR>0.7), robust to all market movements 2006-2025, so we can expect it will perform well in future, too.

Dear @lookman,
can you please share part of code where secgov_load_indicators() functions is called and throws error (with full exception)? Keep in mind that this function is actually load_indicators() function from secgov_indicators.py,and can be imported and used as:

from qnt.data import secgov_load_indicators

or

from qnt.data.secgov_indicators import load_indicators

Kind regards

Dear @auxiliary-snail,

1. Out of sample score is currently calculated every 4-5 days, depending on the internal configuration and currently available data, but the plan is to do the calculation immediately after getting new available data.
2. Yes, when you generate weights for some day, those weights are used with the data on the following day.
3. Yes, it is possible, in your strategy environment you can upload additional files that you would like to use, and then you could load that model directly into your code and use it for generating weights only.

Best regards

Dear @angusslq,

That line is recommendation for the users to not create a strategy that exceeds the timeout limits. What that means is that after submitting, the strategy should ideally compute weights for one trading day in less than 10 minutes. Depending on our available resources, we can also increase that limit somewhat but it is not recommended to create strategies that exceed this limit.

Regards

Dear @lookman,

sorry for late response, we are working on fixing this issue. Meanwhile, if it's applicable, you can try with other functions from secgov module for analysis, like secgov_load_indicators().

@sun-73 it should be ok