Research Tier - Live Algorithms Deployment Limit

Hi Ramesh,

When you upgrade to the Research tier, you can have up to 2 live trading nodes source. Each live trading node can run one algorithm, so you can deploy up to 2 live algorithms simultaneously. If you want to deploy more live algorithms, you would need to upgrade to a higher tier or add more live nodes to your account.

The deployment of algorithms works on a 1 Algorithm - 1 Live Node basis. This means that each algorithm you want to run live needs its own live trading node. We do not support running multiple algorithms on a single live node source.

To deploy a live algorithm, you would need to follow these steps:

1. Open the project that you want to deploy.
2. Click the Deploy Live icon.
3. On the Deploy Live page, click the Brokerage field and then click Paper Trading from the drop-down menu.
4. Click the Node field and then click the live trading node that you want to use from the drop-down menu.
5. (Optional) In the Data Provider section, click Show and change the data provider or add additional providers.

You can find more detailed instructions on how to deploy a live algorithm here.

I hope this helps! Let me know if you have any other questions.

Best,

Thank you for the quick response. 

Also, is it possible to trade multiple assets (Equities, Crypto, Options) with multiple strategies using one single algorithm? if not, please explain why?

Thanks

Hi Ramesh,

Yes, it is possible to trade multiple assets such as Equities, Crypto, and Options within a single algorithm on QuantConnect. QuantConnect supports multi-asset algorithms, allowing you to trade FX, equities, CFD, Futures, and Options in the same algorithm source.

You can add these assets to your algorithm using the AddEquity, AddCrypto, and AddOption methods respectively. Here is an example in Python:

class MyAlgorithm(QCAlgorithm):
    def Initialize(self):
        self.SetStartDate(2022, 1, 1)
        self.SetEndDate(2022, 12, 31)
        self.SetCash(100000)

        # Add assets
        self.equity = self.AddEquity("AAPL", Resolution.Minute)
        self.crypto = self.AddCrypto("BTCUSD", Resolution.Minute)
        self.option = self.AddOption("AAPL", Resolution.Minute)

However, please note that while you can trade multiple assets within a single algorithm, it is not currently possible to use multiple brokerages within the same algorithm source. This means that if you want to trade assets that are only available through certain brokerages, you would need to run separate algorithms for each brokerage.

I hope this helps! Let me know if you have any other questions.

Best,

Ok. That sounds good !

Also, is it possible to implement multiple strategies (Momentum, Mean Reversion, Breakout etc) in one single algorithm ?

Thanks

Hi Ramesh,

Yes, it is possible to implement multiple strategies within a single algorithm on QuantConnect. You can do this by creating different methods for each strategy and calling them in your OnData method. Here is an example in C# from one of the forum discussions:

private void OnData(TradeBars data)
{
    CallTrendFollowingStrategy(data);
    CallMeanReversionStrategy(data);
    CallPairsTradingStrategy(data);
}

// Logic from your Trend Following Algorithm:
private void CallTrendFollowingStrategy(TradeBars bars)
{
    foreach(var bar in bars)
    {
        // Your trend following logic here
    }
}

// Logic from your Mean Reversion Algorithm:
private void CallMeanReversionStrategy(TradeBars bars)
{
    foreach(var bar in bars)
    {
        // Your mean reversion logic here
    }
}

// Logic from your Pairs Trading Algorithm:
private void CallPairsTradingStrategy(TradeBars bars)
{
    // Your pairs trading logic here
}

source

Another approach is to use the Algorithm Framework provided by QuantConnect. This framework allows you to create multiple Alpha Models (strategies) and combine their signals using a Portfolio Construction Model. Here is an example in Python:

class MyAlgorithm(QCAlgorithmFramework):
    def Initialize(self):
        self.SetStartDate(2022, 1, 1)
        self.SetEndDate(2022, 12, 31)
        self.SetCash(100000)

        # Add assets
        self.AddEquity("AAPL", Resolution.Minute)
        self.AddCrypto("BTCUSD", Resolution.Minute)

        # Add Alpha Models
        self.AddAlpha(MyMomentumAlphaModel())
        self.AddAlpha(MyMeanReversionAlphaModel())
        self.AddAlpha(MyBreakoutAlphaModel())

        # Set Portfolio Construction Model
        self.SetPortfolioConstruction(EqualWeightingPortfolioConstructionModel())

source

I hope this helps! Let me know if you have any other questions.

Best,

Thanks for the response !

How do we specify different start dates and resolutions for different assets back testing in the initialize method?

Thanks 

Hi Ramesh,

In QuantConnect, the start and end dates for backtesting are set at the algorithm level, not at the individual asset level. This means that all assets in your algorithm will use the same start and end dates for backtesting. You can set these dates using the SetStartDate and SetEndDate methods in the Initialize method of your algorithm. Here is an example:

def Initialize(self):
    self.SetStartDate(2022, 1, 1)
    self.SetEndDate(2022, 12, 31)
    self.SetCash(100000)

source

As for the resolution, you can specify different resolutions for different assets when you add them to your algorithm. The resolution determines the frequency of data updates for that asset. Here is an example:

def Initialize(self):
    self.SetStartDate(2022, 1, 1)
    self.SetEndDate(2022, 12, 31)
    self.SetCash(100000)

    # Add assets with different resolutions
    self.AddEquity("AAPL", Resolution.Minute)
    self.AddCrypto("BTCUSD", Resolution.Hourly)

source

I hope this helps! Let me know if you have any other questions.

Best,

What would be the criteria for Equity Universe selection for Mean Reversion and ORB Strategies?

Hi Ramesh,

The criteria for universe selection can vary depending on the specific requirements of your Mean Reversion and Opening Range Breakout (ORB) strategies. However, I can provide some general guidelines based on the information available in the knowledge sources.

For a Mean Reversion strategy, one common approach is to select stocks based on their price and trading volume. For example, you might choose to exclude stocks with prices lower than a certain threshold (e.g., $5) and select the ones with the highest dollar traded volume. Here is an example in Python:

# Sort the equities in DollarVolume decendingly
selected = sorted([x for x in coarse if x.Price > 5],
key=lambda x: x.DollarVolume, reverse=True)
symbols = [x.Symbol for x in selected[:self.num_equities]]

source

For an ORB strategy, you might want to select stocks based on their behavior during pre-market/market open. For example, small cap stocks that increase a certain percentage from market open might be candidates for an intraday mean reversion trade. However, this would require real-time or intraday universe selection, which might involve using a custom data feed or adding securities on an ad-hoc basis source.

Please note that these are just examples and the actual universe selection criteria would depend on the specific requirements of your strategies. You might also want to consider factors such as transaction costs, liquidity, and risk management when selecting your universe.

I hope this helps! Let me know if you have any other questions.

Best,