Hi Xiyuan,
Currently we don’t support direct access to historical fundamental data. However, we could implement Rolling Window to store and use fundamental data.
Here is an example using the average PE ratio in past 90 days for fine selection. We build a rolling window for each symbol with a size of 90. Everyday we add the newest PE ratio into the rolling window, and take the average of the rolling window. Notice that in the first 90 days, the rolling window is not filled, so the average is taken from the start date to the current date (you cannot access data before start date). After 90 days, the rolling window only contains data from the most recent 90 days, and works as expected. You might find more help about rolling window here.
You can always change the size of rolling window and the way it is updated for your own purpose. In addition, we are working on historical fundamental data access support. Hope it helps!
class OptimizedHorizontalCoreWave(QCAlgorithm):
def Initialize(self):
self.SetStartDate(2018, 10, 11) # Set Start Date
self.SetEndDate(2019, 10, 10) # Set End Date
self.SetCash(100000) # Set Strategy Cash
self.SetUniverseSelection(FineFundamentalUniverseSelectionModel(self.CoarseSelectionFunction, self.FineSelectionFunction, None, None))
self.historicalPERatio = {}
self.historicalPERatioAverage = {}
self.rollingWindowSize = 90;
# sort the data by daily dollar volume and take the top 'NumberOfSymbols'
def CoarseSelectionFunction(self, coarse):
# sort descending by daily dollar volume
sortedByDollarVolume = sorted(coarse, key=lambda x: x.DollarVolume, reverse=True)
# return the symbol objects of the top entries from our sorted collection
return [ x.Symbol for x in sortedByDollarVolume[:1000] ]
# sort the data by P/E ratio and take the top 'NumberOfSymbolsFine'
def FineSelectionFunction(self, fine):
for x in fine:
if x.Symbol not in self.historicalPERatio.keys():
# Set up rolling window for new ticker
self.historicalPERatio[x.Symbol] = RollingWindow[Decimal](self.rollingWindowSize)
# Add PERatio to rolling window
self.historicalPERatio[x.Symbol].Add(x.ValuationRatios.PERatio)
# Take average of RW
average = 0.0
for pe in self.historicalPERatio[x.Symbol]:
average += pe
average /= self.historicalPERatio[x.Symbol].Count
self.historicalPERatioAverage[x.Symbol] = average
# sort descending by average P/E ratio
sortedByPeRatio = sorted(fine, key=lambda x: self.historicalPERatioAverage[x.Symbol], reverse=True)
# take the top 5 entries from our sorted collection
return [ x.Symbol for x in sortedByPeRatio[:5] ]
Start New Discussion 
Back
Reward Discussion 
1
