def _init_estimation_flow(self):
flow_name = self.user_context.flow_config.get("est_flow_name", "est_flow")
self._estimation_flow = SensorFlow(name=flow_name, data_manager=self.user_context.DM)
# factor date = yesterday
self._estimation_flow.add_next_step2(name="factor_as_of_date",
sensor=GetDate,
kwds={'offset': 1}
)
# module 4. 确定对Risk数据进行数据清洗的集合
self._estimation_flow.add_next_step2(name="riskPool",
sensor=GetPool,
call=None,
# 这里用factor_as_of_date
input_var=[f"{flow_name}.factor_as_of_date.date"],
kwds={"pool_name": self.user_context.pool_name,
"benchmark_weight": "weight_index_500"
})
# module 11. 保存
self._estimation_flow.add_next_step2(
name="saveToNpy_pool",
sensor=SaveToBundleSensor,
call=None,
input_var=[f"{flow_name}.factor_as_of_date.date",
f"{flow_name}.riskPool.pool"],
kwds={
'bundle': self.user_context.config.base.data_bundle_path,
'type': "pool",
'name': "pool_01_final"
}
)
def _init_prepare_flow(self):
self._prepare_flow = SensorFlow(name="prepare_flow",
data_manager=self.user_context.DM)
# module 1. 确定Alpha因子清单
self._prepare_flow.add_next_step2(
name="alphaFactorList",
sensor=GetFactorList,
kwds={"factor_list": self.user_context.alphaFactorDataFrame})
def _init_store_flow(self):
est_flow = self.user_context.flow_config.get("est_flow_name",
"estimation_flow")
self._store_flow = SensorFlow(name="store_flow",
data_manager=self.user_context.DM)
self._store_flow.add_next_step2(
name="store",
sensor=Store,
call=None,
input_var=[
"store_flow.store.factorReturn", "store_flow.store.factorName",
"store_flow.store.dateList", f"{est_flow}.OLS.coefficient",
"prepare_flow.alphaFactorList.factorList"
],
alias={
"store_flow.store.factorName": "col_names",
"prepare_flow.alphaFactorList.factorList": "factorName"
},
kwds={})
def _init_prediction_stock(self):
est_flow = self.user_context.flow_config.get("est_flow_name",
"estimation_flow")
pred_flow = self.user_context.flow_config.get("pred_flow_name",
"prediction_stock")
# silent = (not self.user_context.debug) if hasattr(self.user_context, "debug") else True
self._prediction_stock = SensorFlow(pred_flow,
data_manager=self.user_context.DM)
# module 2. 预测factor_return
self._prediction_stock.add_next_step2(
name="predictionFactorReturn",
sensor=PredictionFactorReturn,
input_var=[
"store_flow.store.factorReturn", "store_flow.store.factorName"
],
kwds={"rolling_window": self.user_context.rolling_window},
silent=False)
# module 3. 预测因子协方差矩阵
self._prediction_stock.add_next_step2(
name="predictionFactorCovariance",
sensor=PredictionFactorCovariance,
input_var=[
"store_flow.store.factorReturn", "store_flow.store.factorName"
],
kwds={"rolling_window": self.user_context.rolling_window},
silent=False)
# 取昨日因子数据
self._prediction_stock.add_next_step2(name="factor_as_of_date",
sensor=GetDate,
kwds={'offset': 1})
factorList = {}
for k in self.user_context.alphaFactorDataFrame.factor_dataFrame.factor:
factorList[k + "_f1"] = FACTOR_STYLE.SECTOR
# module 7. 取alpha数据
self._prediction_stock.add_next_step2(
name="orthogonalization",
sensor=GetFactorData,
call=None,
input_var=[f"{pred_flow}.factor_as_of_date.date"],
kwds={"factorList": factorList})
# module 7. 预测股票收益
self._prediction_stock.add_next_step2(
name="predictionStockReturn",
sensor=PredictionStockReturn,
input_var=[
f"{pred_flow}.predictionFactorReturn.factorReturn",
f"{pred_flow}.orthogonalization.exposure",
f"{est_flow}.OLS.model"
],
kwds={},
silent=False)
self._prediction_stock.add_next_step2(
name="saveToNpy_return",
sensor=SaveToBundleSensor,
call=None,
input_var=[
f"{pred_flow}.factor_as_of_date.date",
f"{pred_flow}.predictionStockReturn.stockReturn"
],
kwds={
'bundle': self.user_context.config.base.data_bundle_path,
'suffix': 'f1',
'type': "return",
'name': "predicted_stock_return"
})
def _init_estimation_flow(self):
flow_name = self.user_context.flow_config.get("est_flow_name", "est_flow")
self._estimation_flow = SensorFlow(name=flow_name, data_manager=self.user_context.DM)
# factor date
self._estimation_flow.add_next_step2(name="factor_as_of_date",
sensor=GetDate,
kwds={'offset': self.user_context.forward_period + 2}
)
# module 4. 确定对Risk数据进行数据清洗的集合
self._estimation_flow.add_next_step2(name="riskPool",
sensor=GetPool,
call=None,
# 这里用factor_as_of_date
input_var=[f"{flow_name}.factor_as_of_date.date"],
kwds={"pool_name": self.user_context.pool_name})
factorList = {}
for k in self.user_context.riskFactorDataFrame.factor_dataFrame.factor:
factorList[k] = FACTOR_STYLE.SECTOR if k.startswith("industry") else FACTOR_STYLE.RISK
# module 6. 取risk数据
self._estimation_flow.add_next_step2(name="riskFactorData",
sensor=GetFactorData,
call=None,
input_var=[f"{flow_name}.riskPool.pool",
f"{flow_name}.factor_as_of_date.date"
],
kwds={"factorList": factorList,
"data_process_methods": {
FACTOR_STYLE.SECTOR: [],
FACTOR_STYLE.RISK: [
DataProcessing.do_process_extremum_winsorize,
DataProcessing.do_z_score_processing
]
}},
silent=False)
try:
open_price_type = self.user_context.est_open_price
except Exception as e:
user_log.warning("no est_open_price in config file")
open_price_type = "open_aft"
try:
close_price_type = self.user_context.est_close_price
except Exception as e:
user_log.warning("no est_close_price in config file")
close_price_type = "open_aft"
user_log.info("est open_price_type is : " + open_price_type)
user_log.info("est close_price_type is : " + close_price_type)
# module 8. 取return数据
self._estimation_flow.add_next_step2(name="returnData",
sensor=GetReturnData, call=None,
input_var=[f"{flow_name}.riskFactorData.exposure",
f"{flow_name}.riskPool.pool",
f"{flow_name}.factor_as_of_date.date"
],
alias={
f"{flow_name}.riskFactorData.exposure": "neutralize_matrix",
},
kwds={"data_process_methods": [
DataProcessing.do_process_extremum_winsorize,
DataProcessing.neutrialize],
"n": self.user_context.forward_period,
"open_price_type": open_price_type,
"close_price_type": close_price_type
},
silent=True)
self._estimation_flow.add_next_step2(
name="saveToNpy_return",
sensor=SaveToBundleSensor,
call=None,
input_var=[f"{flow_name}.factor_as_of_date.date",
f"{flow_name}.returnData.stockReturn"],
kwds={
'bundle': self.user_context.config.base.data_bundle_path,
'suffix': 'f1',
'type': "return",
'name': "forward_return_%d" % self.user_context.forward_period
}
)
def _init_estimation_flow(self):
flow_name = self.user_context.flow_config.get("est_flow_name",
"est_flow")
risk_flow_name = self.user_context.flow_config.get(
"forward_return_flow_name", "flow_forward_return")
self._estimation_flow = SensorFlow(name=flow_name,
data_manager=self.user_context.DM)
# factor date = forward_period + 2
self._estimation_flow.add_next_step2(name="factor_as_of_date",
sensor=GetDate,
kwds={'offset': 1})
# module 4. 确定对Risk数据进行数据清洗的集合
self._estimation_flow.add_next_step2(
name="riskPool",
sensor=GetPool,
call=None,
# 这里用factor_as_of_date
input_var=[f"{flow_name}.factor_as_of_date.date"],
kwds={"pool_name": self.user_context.pool_name})
factorList = {}
for k in self.user_context.alphaFactorDataFrame.factor_dataFrame.factor:
factorList[k] = FACTOR_STYLE.ALPHA
# module 7. 取alpha数据
self._estimation_flow.add_next_step2(
name="alphaFactorData",
sensor=GetFactorData,
call=None,
input_var=[
f"{risk_flow_name}.riskFactorData.exposure",
f"{flow_name}.riskPool.pool",
f"{flow_name}.factor_as_of_date.date"
],
kwds={
"factorList": factorList,
"data_process_methods": {
FACTOR_STYLE.ALPHA: [
DataProcessing.do_process_extremum_winsorize,
DataProcessing.do_z_score_processing,
DataProcessing.neutrialize
]
}
},
silent=True)
self._estimation_flow.add_next_step2(
name="orthogonalization",
sensor=Orthogonalization_Schmidt,
call=None,
input_var=[
f"{flow_name}.alphaFactorData.exposure",
f"{flow_name}.alphaFactorData.factorName",
f"{flow_name}.riskPool.pool"
],
kwds={})
self._estimation_flow.add_next_step2(
name="saveToNpy_alpha",
sensor=SaveToBundleSensor,
call=None,
input_var=[
f"{flow_name}.factor_as_of_date.date",
f"{flow_name}.orthogonalization.exposure",
f"{flow_name}.orthogonalization.factorName"
],
kwds={
'bundle': self.user_context.config.base.data_bundle_path,
'type': "factor",
'suffix': 'f1'
})