<![CDATA[Strategy help]]>http://quantiacs.com/community/category/2RSS for NodeSat, 11 Apr 2026 08:56:20 GMTSat, 11 Apr 2026 08:33:03 GMT60<![CDATA[scritchy scratchy]]>thanks for sharing this! i had the same question about converting stateful strategies to single pass for submission. the scritchy scratchy guide helped me understand the difference between multi and single backtest modes - hope the quantiacs team releases more documentation on this soon.

]]>http://quantiacs.com/community/topic/783/scritchy-scratchyhttp://quantiacs.com/community/topic/783/scritchy-scratchySat, 11 Apr 2026 08:33:03 GMT<![CDATA[Is there a way to submit a strategy via the API?]]>@svyable Hi,
sorry for late answer, no we don't provide that option, but we will think about adding it in future.

]]>http://quantiacs.com/community/topic/761/is-there-a-way-to-submit-a-strategy-via-the-apihttp://quantiacs.com/community/topic/761/is-there-a-way-to-submit-a-strategy-via-the-apiFri, 13 Feb 2026 09:29:41 GMT<![CDATA[Q24 Example Failed to Run]]>@tim-hub Hi,

the template processing has not finished yet, it is on day by day processing (currently beginning of 2023). The reason of the error is that helper function tries to compare stat after 2022-08-31, which is not possible if processing date is 2020-01-01 for example. It doesn't affect the output (weights), and now you don't see this error because processing date is after 2022-08-31. If you had cloned the template and run it, the error would not have occurred.

]]>http://quantiacs.com/community/topic/722/q24-example-failed-to-runhttp://quantiacs.com/community/topic/722/q24-example-failed-to-runWed, 03 Dec 2025 11:10:49 GMT<![CDATA[Strategy filtered after a few days]]>@sun-73 it should be ok

]]>http://quantiacs.com/community/topic/631/strategy-filtered-after-a-few-dayshttp://quantiacs.com/community/topic/631/strategy-filtered-after-a-few-daysTue, 14 Jan 2025 07:27:58 GMT<![CDATA[How to submit stateful long short]]>@newbiequant96 Hi, the template is a "working code" still to be finalized and published among the templates in the account area, however the logic behind is strictly multi-pass and a conversion to single pass is not really so straightforward.

]]>http://quantiacs.com/community/topic/590/how-to-submit-stateful-long-shorthttp://quantiacs.com/community/topic/590/how-to-submit-stateful-long-shortMon, 10 Jun 2024 10:14:10 GMT<![CDATA[Os period is not updated]]>@commanderangle Dear commanderangle,

Your strategies are processed in a correct manner, but the reason why you see 0 out-of-sample score is due to the fact that your strategies generate zero weights for all assets for out-of-sample time period. You can check your weights for any strategy by downloading them. There is a download button in the submission logs section.

Regards

]]>http://quantiacs.com/community/topic/589/os-period-is-not-updatedhttp://quantiacs.com/community/topic/589/os-period-is-not-updatedMon, 10 Jun 2024 07:19:37 GMT<![CDATA[Optimization speed in python]]>@dark-pidgeot Hi, the question is a bit too generic, what do you need exactly to speed up?

]]>http://quantiacs.com/community/topic/553/optimization-speed-in-pythonhttp://quantiacs.com/community/topic/553/optimization-speed-in-pythonTue, 09 Apr 2024 16:36:32 GMT<![CDATA[Missed Call to write output]]>I finally resolved the issue, after lots of struggle. The custom layers, custom loss function and the function had to be serialized and deserialized correctly in order to save the architecture and weights as Json, rather than in a dictionary, like is suggested for pytorch in the Neural netowork template.
It seems Pytorch is way more user friendly when it comes to saving and loading models.

]]>http://quantiacs.com/community/topic/522/missed-call-to-write-outputhttp://quantiacs.com/community/topic/522/missed-call-to-write-outputSat, 06 Apr 2024 01:27:27 GMT<![CDATA[progress check froze]]>@support Hello,

got it, thanks for the reply,

]]>http://quantiacs.com/community/topic/539/progress-check-frozehttp://quantiacs.com/community/topic/539/progress-check-frozeMon, 18 Mar 2024 11:52:43 GMT<![CDATA[TypeError: __init__() got an unexpected keyword argument 'max_value']]>@illustrious-felice
Hello, sorry for late reply, can you please let us know about the progressbar2 version installed from environment you are currently using? Be sure that "progressbar2" package is installed and not "progressbar".

Best regards,

]]>http://quantiacs.com/community/topic/538/typeerror-__init__-got-an-unexpected-keyword-argument-max_valuehttp://quantiacs.com/community/topic/538/typeerror-__init__-got-an-unexpected-keyword-argument-max_valueMon, 18 Mar 2024 08:28:09 GMT<![CDATA[How to get NASDAQ100 index weights?]]>@buyers_are_back Hi, more info of course can be found here:

https://indexes.nasdaqomx.com/docs/Nasdaq_Index_Weight_Calculations.pdf

and here:

https://indexes.nasdaq.com/docs/Methodology_NDX.pdf

Pages 4 and 5 give more details on the weights and caps.

For example, no security weight may exceed 14% of the index value.

]]>http://quantiacs.com/community/topic/536/how-to-get-nasdaq100-index-weightshttp://quantiacs.com/community/topic/536/how-to-get-nasdaq100-index-weightsTue, 12 Mar 2024 20:15:57 GMT<![CDATA[RuntimeError: expand(torch.DoubleTensor{[694, 6]}, size=[694]): the number of sizes provided (1) must be greater or equal to the number of dimensions in the tensor (2)]]>@support Thank you so much. I have resolved this error

]]>http://quantiacs.com/community/topic/529/runtimeerror-expand-torch-doubletensor-694-6-size-694-the-number-of-sizes-provided-1-must-be-greater-or-equal-to-the-number-of-dimensions-in-the-tensor-2http://quantiacs.com/community/topic/529/runtimeerror-expand-torch-doubletensor-694-6-size-694-the-number-of-sizes-provided-1-must-be-greater-or-equal-to-the-number-of-dimensions-in-the-tensor-2Sun, 10 Mar 2024 15:22:08 GMT<![CDATA[backtest_ml has too long a run time]]>@illustrious-felice

Incorporating seed initialization into your PyTorch code ensures reproducibility by making the random number generation predictable. This involves setting seeds for the PyTorch engine, NumPy, and the Python random module if you're using it. Below, I'll show you how to integrate seed initialization into your existing code. Remember, while this can make your experiments more reproducible, it does not guarantee identical results across different hardware or PyTorch versions due to the inherent nondeterminism in some GPU operations.

import xarray as xr # xarray for data manipulation import qnt.data as qndata # functions for loading data import qnt.backtester as qnbt # built-in backtester import qnt.ta as qnta # technical analysis library import numpy as np import pandas as pd import torch from torch import nn, optim import random # Seed initialization function def set_seed(seed_value=42): """Set seed for reproducibility.""" random.seed(seed_value) np.random.seed(seed_value) torch.manual_seed(seed_value) torch.cuda.manual_seed(seed_value) torch.cuda.manual_seed_all(seed_value) # if you are using multi-GPU. torch.backends.cudnn.deterministic = True torch.backends.cudnn.benchmark = False # Set the seed for reproducibility set_seed(42) asset_name_all = ['NAS:AAPL', 'NAS:AMZN', 'NAS:MSFT'] class LSTM(nn.Module): """ Class to define our LSTM network. """ def __init__(self, input_dim=3, hidden_layers=64): super(LSTM, self).__init__() self.hidden_layers = hidden_layers self.lstm1 = nn.LSTMCell(input_dim, self.hidden_layers) self.lstm2 = nn.LSTMCell(self.hidden_layers, self.hidden_layers) self.linear = nn.Linear(self.hidden_layers, 1) def forward(self, y, future_preds=0): outputs = [] n_samples = y.size(0) h_t = torch.zeros(n_samples, self.hidden_layers, dtype=torch.float32) c_t = torch.zeros(n_samples, self.hidden_layers, dtype=torch.float32) h_t2 = torch.zeros(n_samples, self.hidden_layers, dtype=torch.float32) c_t2 = torch.zeros(n_samples, self.hidden_layers, dtype=torch.float32) for time_step in range(y.size(1)): x_t = y[:, time_step, :] # Ensure x_t is [batch, input_dim] h_t, c_t = self.lstm1(x_t, (h_t, c_t)) h_t2, c_t2 = self.lstm2(h_t, (h_t2, c_t2)) output = self.linear(h_t2) outputs.append(output.unsqueeze(1)) outputs = torch.cat(outputs, dim=1).squeeze(-1) return outputs def get_model(): model = LSTM(input_dim=3) return model def get_features(data): close_price = data.sel(field="close").ffill('time').bfill('time').fillna(1) open_price = data.sel(field="open").ffill('time').bfill('time').fillna(1) high_price = data.sel(field="high").ffill('time').bfill('time').fillna(1) log_close = np.log(close_price) log_open = np.log(open_price) features = xr.concat([log_close, log_open, high_price], "feature") return features def get_target_classes(data): price_current = data.sel(field='close') price_future = qnta.shift(price_current, -1) class_positive = 1 # prices goes up class_negative = 0 # price goes down target_price_up = xr.where(price_future > price_current, class_positive, class_negative) return target_price_up def load_data(period): return qndata.stocks.load_ndx_data(tail=period, assets=asset_name_all) def train_model(data): features_all = get_features(data) target_all = get_target_classes(data) models = dict() for asset_name in asset_name_all: model = get_model() target_cur = target_all.sel(asset=asset_name).dropna('time', 'any') features_cur = features_all.sel(asset=asset_name).dropna('time', 'any') target_for_learn_df, feature_for_learn_df = xr.align(target_cur, features_cur, join='inner') criterion = nn.MSELoss() optimiser = optim.LBFGS(model.parameters(), lr=0.08) epochs = 1 for i in range(epochs): def closure(): optimiser.zero_grad() feature_data = feature_for_learn_df.transpose('time', 'feature').values in_ = torch.tensor(feature_data, dtype=torch.float32).unsqueeze(0) out = model(in_) target = torch.zeros(1, len(target_for_learn_df.values)) target[0, :] = torch.tensor(np.array(target_for_learn_df.values)) loss = criterion(out, target) loss.backward() return loss optimiser.step(closure) models[asset_name] = model return models def predict(models, data): weights = xr.zeros_like(data.sel(field='close')) for asset_name in asset_name_all: features_all = get_features(data) features_cur = features_all.sel(asset=asset_name).dropna('time', 'any') if len(features_cur.time) < 1: continue feature_data = features_cur.transpose('time', 'feature').values in_ = torch.tensor(feature_data, dtype=torch.float32).unsqueeze(0) out = models[asset_name](in_) prediction = out.detach()[0] weights.loc[dict(asset=asset_name, time=features_cur.time.values)] = prediction return weights weights = qnbt.backtest_ml( load_data=load_data, train=train_model, predict=predict, train_period=55, retrain_interval=55, retrain_interval_after_submit=1, predict_each_day=False, competition_type='stocks_nasdaq100', lookback_period=55, start_date='2024-01-01', build_plots=True )

I think I won't be available next week. If you have any more questions, don’t expect an answer from me next week.

]]>http://quantiacs.com/community/topic/528/backtest_ml-has-too-long-a-run-timehttp://quantiacs.com/community/topic/528/backtest_ml-has-too-long-a-run-timeFri, 08 Mar 2024 16:26:54 GMT<![CDATA[Errors when I save the isssus parameters of my optimization in the json file]]>@support Thank you for your advise, it's ok

]]>http://quantiacs.com/community/topic/523/errors-when-i-save-the-isssus-parameters-of-my-optimization-in-the-json-filehttp://quantiacs.com/community/topic/523/errors-when-i-save-the-isssus-parameters-of-my-optimization-in-the-json-fileThu, 22 Feb 2024 05:40:02 GMT<![CDATA[What is forward looking and why it's effective badly to strategy?]]>@illustrious-felice Hi, that is not automatic as all trades are punished by slippage, and that is a subtraction from profits irrespective on the sign of the weights.

]]>http://quantiacs.com/community/topic/515/what-is-forward-looking-and-why-it-s-effective-badly-to-strategyhttp://quantiacs.com/community/topic/515/what-is-forward-looking-and-why-it-s-effective-badly-to-strategyFri, 16 Feb 2024 16:35:54 GMT<![CDATA[Please create the program "Quantiacs Tips"]]>@support Thank you for your feedback. I also hope Quantiacs updates new strategy examples on how to use technical analysis (besides sma, trix_ema, atr_lwma,...), and strategies on using ML/DL models effectively (not an example that strategy forward-looking),...

Hopefully in the future Quantiacs will release new data sets such as news, sentiment, macro, options,... Create new contests that allow merging strategies to build portfolios,...

Hopefully, Quantiacs will continue to grow. Sincere thanks to Quantiacs for creating extremely high-quality contests.

]]>http://quantiacs.com/community/topic/513/please-create-the-program-quantiacs-tipshttp://quantiacs.com/community/topic/513/please-create-the-program-quantiacs-tipsTue, 13 Feb 2024 12:02:12 GMT<![CDATA[How to select and combine strategies to optimize your portfolio]]>@support. Thank you. So as I understand it, I will give higher weight to strategies with lower correlation and vice versa, right? According to your answer, I understand that I can also give high weight to low-volatility strategies and vice versa. So what about equal risk portfolio? In your opinion, is this an effective way to optimize your portfolio?

]]>http://quantiacs.com/community/topic/505/how-to-select-and-combine-strategies-to-optimize-your-portfoliohttp://quantiacs.com/community/topic/505/how-to-select-and-combine-strategies-to-optimize-your-portfolioMon, 22 Jan 2024 04:47:38 GMT<![CDATA[Technique to reduce max_drawdown]]>@magenta-kabuto Thank you very much for your advice. I will research to apply your suggestions to the algorithm

]]>http://quantiacs.com/community/topic/491/technique-to-reduce-max_drawdownhttp://quantiacs.com/community/topic/491/technique-to-reduce-max_drawdownThu, 28 Dec 2023 14:52:42 GMT<![CDATA[Checking of strategies for Q20 takes two weeks]]>@algotime Dear Algotime, all three strategies will participate in the contest. We will update the contest leaderboard once all eligible strategies have finished processing. Thank you for your patience.

]]>http://quantiacs.com/community/topic/440/checking-of-strategies-for-q20-takes-two-weekshttp://quantiacs.com/community/topic/440/checking-of-strategies-for-q20-takes-two-weeksMon, 09 Oct 2023 14:00:42 GMT<![CDATA[How to load data to work with Multi-backtesting_ml]]>@cyan-gloom

Hello. The provided code is insufficient to understand the problem.

I assume that a certain function might not be returning the required value (for instance, the function where your model is being created).

I recommend that you check all return values of functions, using tools like display or print. Then, compare them with what is returned in properly working examples.

The state allows you to use data from previous iterations. You can find an example here:
https://github.com/quantiacs/toolbox/blob/2f4c42e33c7ce789dfad5d170444fd542e28c8ae/qnt/examples/004-strategy-futures-multipass-stateful.py

]]>http://quantiacs.com/community/topic/425/how-to-load-data-to-work-with-multi-backtesting_mlhttp://quantiacs.com/community/topic/425/how-to-load-data-to-work-with-multi-backtesting_mlTue, 05 Sep 2023 08:43:25 GMT<![CDATA[Error Q20 output missing when submitting]]>@theflyingdutchman

it can help you. See topic 3) Not enough bid information.

https://github.com/quantiacs/strategy-q20-nasdaq100-quick-start/blob/master/strategy.ipynb

def get_enough_bid_for(data_, weights_): time_traded = weights_.time[abs(weights_).fillna(0).sum('asset') > 0] is_strategy_traded = len(time_traded) if is_strategy_traded: return xr.where(weights_.time < time_traded.min(), data_.sel(field="is_liquid"), weights_) return weights_ weights_new = get_enough_bid_for(data, weights) weights_new = weights_new.sel(time=slice("2006-01-01",None)) ]]>http://quantiacs.com/community/topic/411/error-q20-output-missing-when-submittinghttp://quantiacs.com/community/topic/411/error-q20-output-missing-when-submittingSun, 13 Aug 2023 10:54:49 GMT<![CDATA[Multi-pass Backtesting]]>@eddiee

Hello.

This code looks to the future.
It is needed to train the model.
Pay attention to the name of the variable.

]]>http://quantiacs.com/community/topic/415/multi-pass-backtestinghttp://quantiacs.com/community/topic/415/multi-pass-backtestingThu, 10 Aug 2023 06:51:03 GMT<![CDATA[Why Sharp ratios is not inverted ?]]>@support
Thanks a lot !

]]>http://quantiacs.com/community/topic/392/why-sharp-ratios-is-not-invertedhttp://quantiacs.com/community/topic/392/why-sharp-ratios-is-not-invertedSun, 28 May 2023 05:32:02 GMT<![CDATA[combinations of strategy]]>@support
Thanks
I got it !

]]>http://quantiacs.com/community/topic/391/combinations-of-strategyhttp://quantiacs.com/community/topic/391/combinations-of-strategyWed, 24 May 2023 07:49:56 GMT<![CDATA[Backtesting]]>@stefanm Thank you!

]]>http://quantiacs.com/community/topic/372/backtestinghttp://quantiacs.com/community/topic/372/backtestingSat, 01 Apr 2023 17:51:09 GMT<![CDATA[Fundamental Data: Periodic indicators & Instant indicators]]>@johback

Hello

More examples are here https://github.com/quantiacs/toolbox/blob/main/qnt/tests/test_fundamental_data.py

This is a simple example.

import qnt.data as qndata import datetime as dt import qnt.data.secgov_indicators import qnt.data as qndata import qnt.stats as qns assets = qndata.stocks.load_ndx_list(tail=dt.timedelta(days=5 * 365)) assets_names = [i["id"] for i in assets] data = qndata.stocks.load_ndx_data(tail=dt.timedelta(days=5 * 365), dims=("time", "field", "asset"), assets=assets_names, forward_order=True) facts_names = ['operating_expense'] # 'assets', 'liabilities', 'ivestment_short_term' and other fundamental_data = qnt.data.secgov_load_indicators(assets, time_coord=data.time, standard_indicators=facts_names) # Operating expenses include marketing, noncapitalized R&D, # travel and entertainment, office supply, rent, salary, cogs... weights = fundamental_data.sel(field='operating_expense') is_liquid = data.sel(field="is_liquid") weights = weights * is_liquid # calc stats stats = qns.calc_stat(data, weights.sel(time=slice("2006-01-01", None))) display(stats.to_pandas().tail()) # graph performance = stats.to_pandas()["equity"] import qnt.graph as qngraph qngraph.make_plot_filled(performance.index, performance, name="PnL (Equity)", type="log") ]]>http://quantiacs.com/community/topic/353/fundamental-data-periodic-indicators-instant-indicatorshttp://quantiacs.com/community/topic/353/fundamental-data-periodic-indicators-instant-indicatorsFri, 27 Jan 2023 09:33:17 GMT