def __init__(self, dataSeries, period, numStdDev, maxLen=None):
self.__sma = ma.SMA(dataSeries, period, maxLen=maxLen)
self.__stdDev = stats.StdDev(dataSeries, period, maxLen=maxLen)
self.__upperBand = dataseries.SequenceDataSeries(maxLen)
self.__lowerBand = dataseries.SequenceDataSeries(maxLen)
self.__numStdDev = numStdDev
# It is important to subscribe after sma and stddev since we'll use those values.
dataSeries.getNewValueEvent().subscribe(self.__onNewValue)
def __init__(self, maxLen=None):
super(analyzer_skysense, self).__init__()
self.__netReturns = dataseries.SequenceDataSeries(maxLen=maxLen)
self.__cumReturns = dataseries.SequenceDataSeries(maxLen=maxLen)
self.__benchmarkDailyReturns = dataseries.SequenceDataSeries(
maxLen=maxLen)
self.__benchmarkCumuReturns = dataseries.SequenceDataSeries(
maxLen=maxLen)
self.__benchmarkList = []
self.__dailyPeriodAnalysis = {
'Date': [],
'profit': [],
'profitratio': []
}
self.__presentEquity = None
self.__lastEquity = None
self.__presentBenchmark = None
self.__lastBenchmark = None
self.__strat = None
self.__totalSlippage = 0
self.__momentPrice = 0
self.__actionList = ["BUY", "BUY_TO_COVER", "SELL", "SELL_SHORT"]
self.__tradesListDict = {
'type': [],
'date': [],
'price': [],
'shares': [],
'return': [],
'returnrate': [],
'instrument': [],
'position': [],
'cost': [],
'datetime': []
}
self.__long = 0
self.__short = 0
self.__countTrades = 0
self.__positionsList = []
self.__dateTimeList = []
self.__equityList = []
self.__cashList = []
self.__barsList = []
self.__buyPricesList = []
self.__buySharesList = []
self.__sellShortPricesList = []
self.__sellShortSharesList = []
self.__holdingCostList = []
self.__buyPricesDict = {}
self.__buySharesDict = {}
self.__sellShortPricesDict = {}
self.__sellShortSharesDict = {}
self.__positionsListByOrder = [{}]
self.__orderDatetimeList = []
def datetime_aligned(ds1, ds2, maxLen=None):
"""
Returns two dataseries that exhibit only those values whose datetimes are in both dataseries.
:param ds1: A DataSeries instance.
:type ds1: :class:`DataSeries`.
:param ds2: A DataSeries instance.
:type ds2: :class:`DataSeries`.
:param maxLen: The maximum number of values to hold for the returned :class:`DataSeries`.
Once a bounded length is full, when new items are added, a corresponding number of items are discarded from the
opposite end. If None then dataseries.DEFAULT_MAX_LEN is used.
:type maxLen: int.
"""
aligned1 = dataseries.SequenceDataSeries(maxLen)
aligned2 = dataseries.SequenceDataSeries(maxLen)
Syncer(ds1, ds2, aligned1, aligned2)
return (aligned1, aligned2)
def getBenchmarkCumuReturns(self):
cumuReturns = dataseries.SequenceDataSeries(maxLen=None)
dateList = self.getDateTimeList()
benchmarkList = self.getBenchmarkList()
assert len(dateList) == len(benchmarkList)
for i in range(0, len(benchmarkList)):
cumuReturns.appendWithDateTime(dateTime=dateList[i],
value=(benchmarkList[i] - benchmarkList[0]) / benchmarkList[0])
return cumuReturns
def __init__(self, dataSeries, fastEMA, slowEMA, signalEMA, maxLen=None):
assert (fastEMA > 0)
assert (slowEMA > 0)
assert (fastEMA < slowEMA)
assert (signalEMA > 0)
super(MACD, self).__init__(maxLen)
# We need to skip some values when calculating the fast EMA in order for both EMA
# to calculate their first values at the same time.
# I'M FORCING THIS BEHAVIOUR ONLY TO MAKE THIS FITLER MATCH TA-Lib MACD VALUES.
self.__fastEMASkip = slowEMA - fastEMA
self.__fastEMAWindow = ma.EMAEventWindow(fastEMA)
self.__slowEMAWindow = ma.EMAEventWindow(slowEMA)
self.__signalEMAWindow = ma.EMAEventWindow(signalEMA)
self.__signal = dataseries.SequenceDataSeries(maxLen)
self.__histogram = dataseries.SequenceDataSeries(maxLen)
dataSeries.getNewValueEvent().subscribe(self.__onNewValue)
def __init__(self, maxLen=None):
super(BarDataSeries, self).__init__(maxLen)
self.__openDS = dataseries.SequenceDataSeries(maxLen)
self.__closeDS = dataseries.SequenceDataSeries(maxLen)
self.__highDS = dataseries.SequenceDataSeries(maxLen)
self.__lowDS = dataseries.SequenceDataSeries(maxLen)
self.__volumeDS = dataseries.SequenceDataSeries(maxLen)
self.__adjCloseDS = dataseries.SequenceDataSeries(maxLen)
self.__extraDS = {}
self.__useAdjustedValues = False
def __getOrCreateExtraDS(self, name):
ret = self.__extraDS.get(name)
if ret is None:
ret = dataseries.SequenceDataSeries(self.getMaxLen())
self.__extraDS[name] = ret
return ret
def __init__(self, maxLen=None):
super(Returns, self).__init__()
self.__netReturns = dataseries.SequenceDataSeries(maxLen=maxLen)
self.__cumReturns = dataseries.SequenceDataSeries(maxLen=maxLen)
def createDataSeries(self, key, maxLen):
return dataseries.SequenceDataSeries(maxLen)