def __init__(self):
#self.dbhandler = services.get('dbhandler')
self.dbreader = services.get('dbreader')
self.trader = services.get('trader')
self.config = services.get('configurator')
self.predictor = services.get('predictor')
self.model = None
self.start_date = None
self.end_date = None
def __init__(self):
BaseBackTester.__init__(self)
self.model = services.get('mean_reversion_model')
self.window_size = 10
self.threshold = 1.5
def __init__(self):
self.dbhandler = services.get('dbhandler')
self.dbreader = services.get('dbreader')
self.predictor = services.get('predictor')
self.config = services.get('configurator')
def __init__(self):
BaseBackTester.__init__(self)
self.model = services.get('machine_learning_model')
def __init__(self, wait_sec=5):
self.wait_sec = wait_sec
self.dbwriter = services.get('dbwriter')
self.dbhandler = services.get('dbhandler')
def __init__(self):
self.dbhandler = services.get('dbhandler')
def __init__(self):
self.dbreader = services.get('dbreader')
self.config = services.get('configurator')
self.items = {}
def makeDataSet(self, code, start_date, end_date):
df = services.get('dbreader').loadDataFrame(code, start_date, end_date)
return df
services.register('predictor', Predictors())
services.register('trader', MessTrader())
services.register('mean_reversion_model', MeanReversionModel())
services.register('machine_learning_model', MachineLearningModel())
crawler = DataCrawler()
universe = Portfolio()
portfolio = PortfolioBuilder()
mean_backtester = MeanReversionBackTester()
machine_backtester = MachineLearningBackTester()
#crawler.updateAllCodes()
#crawler.updateAllStockData(1, 2010, 1, 1, 2015, 12, 1, start_index=90)
services.get('configurator').register('start_date', '20150101')
services.get('configurator').register('end_date', '20151031')
services.get('configurator').register('input_column', 'price_adj_close')
services.get('configurator').register('output_column', 'indicator')
services.get('configurator').register('data_limit', 20)
#finder.setTimePeriod('20150101', '20151130')
df_stationarity = portfolio.doStationarityTest('price_close')
df_rank = portfolio.rankStationarity(df_stationarity)
stationarity_codes = portfolio.buildUniverse(df_rank, 'rank', 0.8)
#print(stationarity_codes)
df_machine_result = portfolio.doMachineLearningTest(split_ratio=0.75,
lags_count=5)
df_machine_rank = portfolio.rankMachineLearning(df_machine_result)
machine_codes = portfolio.buildUniverse(df_machine_rank, 'rank', 0.8)
