Back

Symbols and rolling windows

I am recieving the following error:


Runtime Error: One or more errors occurred. Specified cast is not valid.

The goal of my program is to use a rolling window of an input variable and and output variable which here is the price to create a prediction for the price. This is my program:

 

#All appropriate variables are imported before this line
# Demonstration of using coarse and fine universe selection together to filter down a smaller universe of stocks.
class FrameworkAlgorithm(QCAlgorithm):

    def Initialize(self):
        '''Initialise the data and resolution required, as well as the cash and start-end dates for your algorithm. All algorithms must initialized.'''
        self.SetStartDate(2014,1,1)  #Set Start Date
        self.SetEndDate(2015,1,1)    #Set End Date
        self.SetCash(50000)            #Set Strategy Cash
        self.UniverseSettings.Resolution = Resolution.Daily        
        
        self.AddUniverse(self.CoarseSelectionFunction, self.FineSelectionFunction)
        self.AddEquity("SPY", Resolution.Daily)
        self.SetBenchmark("SPY")

        self.__numberOfSymbols = 100
        self.__numberOfSymbolsFine = 30

        self._changes = None
        
        self.Schedule.On(self.DateRules.MonthStart("SPY"), self.TimeRules.AfterMarketOpen("SPY"), Action(self.Rebalancing))
        
        self.splotName = 'Strategy Info'
        sPlot = Chart(self.splotName)
        sPlot.AddSeries(Series('Leverage',  SeriesType.Line, 0))
        self.AddChart(sPlot)
        
        self.historicalPERatio = {}
        self.historicalPricetoCashRatio={}
        #self.historicalTotalDebtEquityRatioGrowth = {}
        self.FCFRatio = {}
        #self.historicalEquityPerShareRatio  = {}
        #self.historicalEVToEBITDA = {}
        self.CFOPerShare = {}
        self.Y = {}
        
        self.rollingWindowSize = 90;
        
        self.averages = { };

    def CoarseSelectionFunction(self, coarse):
        x = list(coarse)
        CoarseWithFundamental = [x for x in coarse if (x.HasFundamentalData) and (float(x.Price) > 5)]
        
        filtered = [x for x in CoarseWithFundamental if x.HasFundamentalData
                                      and x.Volume > 0
                                      and x.Price > 0]
        # sort descending by daily dollar volume
        sortedByDollarVolume = sorted(filtered, key=lambda x: x.DollarVolume, reverse=True)
        #HasFundamentalData = sorted(coarse, key=lambda x: HasFundamentalData, reverse=True)
    
        top = sortedByDollarVolume[:self.__numberOfSymbols]
        # return the symbol objects of the top entries from our sorted collection
        return top

    def FineSelectionFunction(self, fine):
        
        
        self.r = [x for x in fine if x.ValuationRatios.PERatio]
    
        for x in self.r:
            if x.Symbol not in self.historicalPERatio.keys():
                # Set up rolling window for new ticker
                self.historicalPERatio[x.Symbol] = RollingWindow[Decimal](self.rollingWindowSize)
            self.historicalPERatio[x.Symbol].Add(x.ValuationRatios.PERatio) #your job is to find the decimal current version for all the following variables
            #self.prices_y[x.Symbol] = list(history.loc[x.Value]['close'])[1:]
            
        self.r = [x for x in fine if PredictionEngine(x.historicalPERatio, x.Y ).predictionScore() > float(x.Price) or PredictionEngine(x.historicalPERatio , x.Y ).predictionScore() < float(x.Price)]
        #self.a = (x for x in self.r  
        r = self.r
        topFine = sorted(self.r, key=lambda x: x.ValuationRatios.PERatio, reverse = False)
        self.topFine = [x.Symbol for x in topFine]

      
        self.symbols = [i.Symbol for i in topFine]
        self.y = self.History(self.symbols, self.rollingWindowSize, Resolution.Daily)
        for i in self.symbols:
            self.Debug(i.Symbol)
            
            
            #self.Debug(self.symbols)
        
        #[x.Symbol for x in self.r]
        return self.symbols

   def OnData(self, data):
        print("asdf")
        symbols = self.symbols
        for x in symbols:
            if PredictionEngine(x.historicalPERatio,x.y).predictionScore() > float(x.Price):
               
                
                self.SetHoldings(security, 0.1)    
                print("Security")
                self.Debug(security)
                #self.historicalPERatio[x.Symbol]
                self._changes = None
                #self.Debug("asdf")
                # if we have no changes, do nothing
        #if self.changes == None: return
                # liquidate removed securities
        #for security in self._changes.RemovedSecurities:
            #if security.Invested:
                
        #for security in self.symbols:
            #self.SetHoldings(security, 0.1*finalFactor)      
    def Rebalancing(self):
        print("b")
        for security in self.Portfolio.Values:    
            if PredictionEngine(security.historicalPERatio,security.y).predictionScore() < float(security.Price)*1.08 and security.Invested:
                self.Liquidate(security.Symbol)

class PredictionEngine(object):
    def __init__(self):
        self.historicalPERatio = historicalPERatio
        self.y = Y
    def predictionScore(self):
        vector_factor_direction= 0
        scalar = StandardScaler()
        p = 1

        pe = pd.DataFrame(self.historicalPERatio)

        #after this line is where I do all the magical things needed to predict prices

    return predictions

Update Backtest







0

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.


Hi Subarno,

Welcome to QC! Please use the Insert Code Snippet button (37376_1567613965.jpg) to insert code so that it becomes highlighted and easier to read.

I wasn't able to narrow down exactly where the error occurs, but looking at the stack trace, my hunch is that it's in one of your Universe Selection functions. The error itself likely suggests you're trying to cast something into a data type it cannot be cast to, which can happen when you're comparing, say, a string and a float. Because I don't know the return of your PredictionEngine.predictionScore() call, I can't really say what's happening.

In general, I have two pieces of advice:

  1. When submitting code to the community, please boil your problem down into its simplest parts, deleting everything that isn't necessary to produce the error. In putting that work in, you often end up finding the answer to the problem yourself, and if not, you're anyway much more likely to receive a helpful reply from the community.
  2. Keep a mental or physical note of the changes you make to your code so you have some idea of what you were doing when the error started appearing. This can be immensely helpful to debug issues like these. If you know what you changed to cause the error - or even if you just know roughly what you were working on - you can help yourself a lot easier.
0

Hi Subarno,

The universe selection function returns a list of symbols. So, you may need to replace 'top' with '[x.Symbol for x in top]' in your coarse selection function.

Also, please post a backtest in the future so that we can better assist you.

I hope this helps.

0

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.


Update Backtest





0

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.


Loading...

This discussion is closed