def plotPriceAndBeLong(issue, modelStartDate, modelEndDate, valData):
from Code.lib.plot_utils import PlotUtility
plotIt = PlotUtility()
plotTitle = "Close " + str(modelStartDate) + " to " + str(modelEndDate)
plotIt.plot_v2x(valData, plotTitle)
plt.show(block=False)
plotIt.histogram(valData['beLong'], x_label="beLong signal", y_label="Frequency",
title = "beLong distribution for " + issue)
plt.show(block=False)
@author: kruegkj
time_test_modelling.py
Goal: Test and verify in sample and out of sample time splits for dataset
"""
from Code.lib.plot_utils import PlotUtility
from Code.lib.time_utils import TimeUtility
from Code.lib.retrieve_data import DataRetrieve, ComputeTarget
from Code.utilities.stat_tests import stationarity_tests, mean_and_variance
import datetime
from dateutil.relativedelta import relativedelta
from pandas.tseries.offsets import BDay
if __name__ == "__main__":
plotIt = PlotUtility()
timeUtil = TimeUtility()
dSet = DataRetrieve()
ct = ComputeTarget()
doPlot = False
issue = "tlt"
pivotDate = datetime.date(2018, 4, 2)
is_oos_ratio = 2
oos_months = 3
segments = 3
# get segmented dates
isOosDates = timeUtil.is_oos_data_split(issue, pivotDate, is_oos_ratio,
oos_months, segments)
dataLoadStartDate = isOosDates[0]
# Import custom libraries
from Code.lib.plot_utils import PlotUtility
from Code.lib.time_utils import TimeUtility
from Code.lib.retrieve_data import DataRetrieve, ComputeTarget
from Code.lib.retrieve_system_info import TradingSystemUtility
from Code.lib.candle_indicators import CandleIndicators
from Code.lib.transformers import Transformers
from Code.lib.ta_momentum_studies import TALibMomentumStudies
from Code.lib.model_utils import ModelUtility, TimeSeriesSplitImproved
from Code.lib.feature_generator import FeatureGenerator
from Code.lib.config import current_feature, feature_dict
from Code.models import models_utils
from Code.lib.model_algos import AlgoUtility
from Code.lib.tms_utils import TradeRisk
plotIt = PlotUtility()
timeUtil = TimeUtility()
ct = ComputeTarget()
candle_ind = CandleIndicators()
dSet = DataRetrieve()
taLibMomSt = TALibMomentumStudies()
transf = Transformers()
modelUtil = ModelUtility()
featureGen = FeatureGenerator()
dSet = DataRetrieve()
modelAlgo = AlgoUtility()
sysUtil = TradingSystemUtility()
tradeRisk = TradeRisk()
if __name__ == '__main__':
UO signal
feature_dict
"""
col_name = 'ROC_' + str(period)
current_feature['Latest'] = col_name
feature_dict[col_name] = 'Keep'
df[col_name] = ta.ROC(df.Close, period)
return df
if __name__ == "__main__":
from Code.lib.plot_utils import PlotUtility
from Code.lib.retrieve_data import DataRetrieve
plotIt = PlotUtility()
taLibMomSt = TALibMomentumStudies()
dSet = DataRetrieve()
issue = "TLT"
df = dSet.read_issue_data(issue)
lastRow = df.shape[0]
dataLoadEndDate = df.Date[lastRow - 100]
pprint(dataLoadEndDate)
dataLoadStartDate = df.Date[lastRow - 3000]
pprint(dataLoadStartDate)
dataSet = dSet.set_date_range(df, dataLoadStartDate, dataLoadEndDate)
pprint(dataSet.tail(10))
from sklearn.model_selection import TimeSeriesSplit
import matplotlib.pylab as plt
from pandas.tseries.offsets import BDay
import os.path
import pickle
#from pandas.tseries.holiday import USFederalHolidayCalendar
#from pandas.tseries.offsets import CustomBusinessDay
#us_cal = CustomBusinessDay(calendar=USFederalHolidayCalendar())
from sklearn.model_selection import StratifiedShuffleSplit
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier, GradientBoostingClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn import svm
if __name__ == "__main__":
plotIt = PlotUtility()
timeUtil = TimeUtility()
ct = ComputeTarget()
candle_ind = CandleIndicators()
dSet = DataRetrieve()
taLibMomSt = TALibMomentumStudies()
transf = Transformers()
modelUtil = ModelUtility()
featureGen = FeatureGenerator()
issue = "TLT"
# Set IS-OOS parameters
pivotDate = datetime.date(2019, 1, 3)
is_oos_ratio = 2
oos_months = 4
segments = 4
from Code.lib.plot_utils import PlotUtility
#from Code.lib.time_utils import TimeUtility
from Code.lib.retrieve_data import DataRetrieve
from Code.lib.retrieve_system_info import TradingSystemUtility
#from Code.lib.candle_indicators import CandleIndicators
#from Code.lib.transformers import Transformers
#from Code.lib.ta_momentum_studies import TALibMomentumStudies
#from Code.lib.model_utils import ModelUtility, TimeSeriesSplitImproved
#from Code.lib.feature_generator import FeatureGenerator
#from Code.utilities.stat_tests import stationarity_tests
#from Code.lib.config import current_feature, feature_dict
#from Code.models import models_utils
#from Code.lib.model_algos import AlgoUtility
from Code.lib.tms_utils import TradeRisk
plotIt = PlotUtility()
#timeUtil = TimeUtility()
#ct = ComputeTarget()
#taLibMomSt = TALibMomentumStudies()
#transf = Transformers()
#modelUtil = ModelUtility()
#featureGen = FeatureGenerator()
dSet = DataRetrieve()
#modelAlgo = AlgoUtility()
sysUtil = TradingSystemUtility()
tradeRisk = TradeRisk()
if __name__ == '__main__':
# set to existing system name OR set to blank if creating new
if len(sys.argv) < 2:
indices[test_start:test_start + test_size])
else:
for test_start in test_starts:
yield (indices[:test_start],
indices[test_start:test_start + test_size])
if __name__ == "__main__":
from Code.lib.plot_utils import PlotUtility
from Code.lib.retrieve_data import DataRetrieve, ComputeTarget
from Code.lib.time_utils import TimeUtility
from Code.lib.feature_generator import FeatureGenerator
from Code.utilities.stat_tests import stationarity_tests
# from Code.lib.config import current_feature, feature_dict
plotIt = PlotUtility()
ct = ComputeTarget()
dSet = DataRetrieve()
featureGen = FeatureGenerator()
timeUtil = TimeUtility()
modelUtil = ModelUtility()
#timeSeriesImp = TimeSeriesSplitImproved()
dataLoadStartDate = "2014-09-26"
dataLoadEndDate = "2018-04-05"
issue = "TLT"
# Set IS-OOS parameters
pivotDate = datetime.date(2018, 4, 2)
is_oos_ratio = 2
oos_months = 4
segments = 4
