RollingWindow: Coding questions by Sebastian Groß

Sebastian Groß

| Reward Discussion

Hi there,

I try to write a pairs trading strategy based on co-integration. This is supposed to be fun project to learn qc datastructur etc.. Unfortunately I do run into various troubles all the time. Maybe someone could help me with this one. In the current algo, I have created two rolling windows. Data should be only added if both assets are present in the onData dataset. If the Windows are ready, cointegration would be estimated. However, this never happens. Any thoughs on that? Note: the SPY position is only for testing since I am also not able to get any log messages to work.

Update Backtest

Sebastian Groß

Accepted Answer

Update Backtest

Notebook

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.

Gurumeher Sawhney

14.1k ,

Thanks for sending the code! Since the event handler OnData(self, data) was lowercase in the code, QCAlgorithm was actually not able to find the method. As a result, the algorithm did not run any other code after initializing. In OnData there were some minor errors in the conditional statements. It is important to note that IsReady is a property of the RollingWindow, not a callable function. Now the RollingWindow is being populated and the strategy portion of the algorithm is running. The code below should get you started. In the files, I noticed a PortfolioConstructionModel and AlphaModel. That is part of QuantConnect's algorithm framework and if they are used the algorithm will need to extend from QCAlgorithmFramework, not QCAlgorithm. The link below will have more documentation on using the framework you end up going in that direction.

https://www.quantconnect.com/docs/algorithm-framework/overview

class LongShortCointegrationPairs(QCAlgorithm):

    def Initialize(self):

        self.asset1 = 'SPY'

        self.asset2 = 'IWM'

        self.periods = 10

        

        self.SetStartDate(2002, 1, 1)  #Set Start Date

        self.SetEndDate(2005, 12, 31)    #Set End Date

        self.SetCash(10000)           #Set Strategy Cash

        

        for sym in [self.asset1, self.asset2]:

            self.AddEquity(sym, Resolution.Daily)

            

        self.p1 = RollingWindow[float](self.periods)

        self.p2 = RollingWindow[float](self.periods)

        self.weights = None

        

    def OnData(self, data):

        # check if prices for both asset are present

        if data[self.asset1] and data[self.asset2]:

            self.p1.Add(data[self.asset1].Close)

            self.p2.Add(data[self.asset2].Close)

        else:

            return None

            

        if self.p1.IsReady and self.p2.IsReady and self.weights is None:

            self.Debug("enter strategy")

            arr = np.matrix([self.p1[:], self.p2[:]]).T

            self.weights = calculate_cointegration_weights(arr)

            self.SetHoldings("SPY", 1)

Sebastian Groß

29 ,

Hi, thanks for fixing my bugs. I would have not be able to find the lower case by myself. I will most certainly use the Framework more extensively in the future, but am still not familiar enough with all the classes and functions required. Thus I wanted to implemt a more simplistic version.

Sebastian Groß

Update Backtest

Notebook

To unlock posting to the community forums please complete at least 30% of Boot Camp.
You can continue your Boot Camp training progress from the terminal. We hope to see you in the community soon!

QuantConnect Community

Alpha League Competition: $1,000 Weekly Prize Pool

Filter Discussion by Tags

160,513 Quants.

Online Now

RollingWindow: Coding questions

Started By:

Join QuantConnect for Free Sign Up