def testCSPercentileSecurityValueHolderWithGroups(self):
benchmark = SecurityLatestValueHolder(x='close')
groups = SecurityLatestValueHolder(x='ind')
perHolder = CSPercentileSecurityValueHolder(benchmark, groups)
for i in range(len(self.datas['aapl']['close'])):
data = {'aapl': {Factors.CLOSE: self.datas['aapl'][Factors.CLOSE][i],
Factors.OPEN: self.datas['aapl'][Factors.OPEN][i],
'ind': 1.},
'ibm': {Factors.CLOSE: self.datas['ibm'][Factors.CLOSE][i],
Factors.OPEN: self.datas['ibm'][Factors.OPEN][i],
'ind': 1.},
'goog': {Factors.CLOSE: self.datas['goog'][Factors.CLOSE][i],
Factors.OPEN: self.datas['goog'][Factors.OPEN][i],
'ind': 2.},
'baba': {Factors.CLOSE: self.datas['baba'][Factors.CLOSE][i],
Factors.OPEN: self.datas['baba'][Factors.OPEN][i],
'ind': 2.}}
benchmark.push(data)
perHolder.push(data)
benchmarkValues = benchmark.value
groups = {'aapl': 1., 'ibm': 1., 'goog': 2., 'baba': 2.}
expected_rank = pd.Series(benchmarkValues.to_dict()).groupby(groups) \
.transform(lambda x: x.rank().values / len(x))
np.testing.assert_array_almost_equal(expected_rank, perHolder.value.values)
def testCSPercentileSecurityValueHolder(self):
benchmark = SecurityLatestValueHolder(x='close')
perHolder = CSPercentileSecurityValueHolder(benchmark)
for i in range(len(self.datas['aapl']['close'])):
data = {
'aapl': {
Factors.CLOSE: self.datas['aapl'][Factors.CLOSE][i],
Factors.OPEN: self.datas['aapl'][Factors.OPEN][i]
},
'ibm': {
Factors.CLOSE: self.datas['ibm'][Factors.CLOSE][i],
Factors.OPEN: self.datas['ibm'][Factors.OPEN][i]
},
'goog': {
Factors.CLOSE: self.datas['goog'][Factors.CLOSE][i],
Factors.OPEN: self.datas['goog'][Factors.OPEN][i]
},
'baba': {
Factors.CLOSE: self.datas['baba'][Factors.CLOSE][i],
Factors.OPEN: self.datas['baba'][Factors.OPEN][i]
}
}
benchmark.push(data)
perHolder.push(data)
benchmarkValues = benchmark.value
np.testing.assert_array_almost_equal(
benchmarkValues.rank().values / (len(data) - 1),
perHolder.value.values)
def testCSPercentileSecurityValueHolder(self):
percent = 50
benchmark = SecurityLatestValueHolder(dependency='close')
perHolder = CSPercentileSecurityValueHolder(percent, benchmark)
for i in range(len(self.datas['aapl']['close'])):
data = {'aapl': {Factors.CLOSE: self.datas['aapl'][Factors.CLOSE][i],
Factors.OPEN: self.datas['aapl'][Factors.OPEN][i]},
'ibm': {Factors.CLOSE: self.datas['ibm'][Factors.CLOSE][i],
Factors.OPEN: self.datas['ibm'][Factors.OPEN][i]}}
benchmark.push(data)
perHolder.push(data)
benchmarkValues = benchmark.value
np.testing.assert_array_almost_equal(np.percentile(benchmarkValues.values, 50), perHolder.value.values)
def CSQuantiles(x, groups=None):
return CSPercentileSecurityValueHolder(x, groups)
def CSPercentile(percent, dependency):
return CSPercentileSecurityValueHolder(percent, dependency)