def get_random_price(price,
code='rb1905',
tradingDay='20181119',
mu=0,
sigma=0.2,
theta=0.15,
dt=1e-2,
ifprint=False,
weight=0.1,
callback=None):
ou_noise = OrnsteinUhlenbeckActionNoise(mu=np.array(mu))
data = []
tick_pickle = {
'InstrumentID':
str(code),
'UpdateTime':
'',
'UpdateMillisec':
'',
'LastPrice':
'',
'Volume':
0,
'TradingDay':
tradingDay,
'ActionDay':
QA_util_date_str2int(
QA_util_get_last_day(QA_util_date_int2str(tradingDay)))
}
for item in time_index:
x = str(item).split()[2]
if '.' in x:
tick_pickle['UpdateMillisec'] = 500
else:
tick_pickle['UpdateMillisec'] = 0
tick_pickle['UpdateTime'] = x.split('.')[0]
if item.seconds >= 75600:
tick_pickle['ActionDay'] = QA_util_date_str2int(
QA_util_get_last_day(
QA_util_date_int2str(tick_pickle['TradingDay'])))
else:
tick_pickle['ActionDay'] = tick_pickle['TradingDay']
tick_pickle['Volume'] += random.randint(50, 5000)
tick_pickle['LastPrice'] = (ou_noise() + 1) * \
weight * price + (1 - weight) * price
data.append(copy.deepcopy(tick_pickle))
if ifprint:
print(tick_pickle)
if callback is not None:
callback(tick_pickle)
return pd.DataFrame(data)
def QA_fetch_get_sh_margin(date):
"""return shanghai margin data
Arguments:
date {str YYYY-MM-DD} -- date format
'code','name','leveraged_balance','leveraged_buyout','leveraged_payoff','margin_left','margin_sell','margin_repay',
标的证券代码 标的证券简称 本日融资余额(元) leveraged_balance
本日融资买入额(元)leveraged_buyout 本日融资偿还额(元) leveraged_payoff
本日融券余量margin_left 本日融券卖出量 margin_sell 本日融券偿还量 margin_repay
Returns:
pandas.DataFrame -- res for margin data
"""
if date in trade_date_sse:
data = pd.read_excel(_sh_url.format(QA_util_date_str2int(date)),
1).assign(date=date).assign(sse='sh')
data.columns = [
'code', 'name', 'leveraged_balance', 'leveraged_buyout',
'leveraged_payoff', 'margin_left', 'margin_sell', 'margin_repay',
'date', 'sse'
]
data.code = data.code.apply(lambda x: str(x)[0:6])
return data
else:
pass
def QA_fetch_get_sh_margin(date):
"""return shanghai margin data
Arguments:
date {str YYYY-MM-DD} -- date format
Returns:
pandas.DataFrame -- res for margin data
"""
if date in trade_date_sse:
data= pd.read_excel(_sh_url.format(QA_util_date_str2int
(date)), 1).assign(date=date).assign(sse='sh')
data.columns=['code','name','leveraged_balance','leveraged_buyout','leveraged_payoff','margin_left','margin_sell','margin_repay','date','sse']
return data
else:
pass
def get_random_price(price,
code='rb1905',
tradingDay='20181119',
mu=0,
sigma=0.2,
theta=0.15,
dt=1e-2,
ifprint=False,
weight=0.1,
market_type=None):
ou_noise = OrnsteinUhlenbeckActionNoise(mu=np.array(mu))
data = []
tick_pickle = {
'InstrumentID':
str(code),
'UpdateTime':
'',
'UpdateMillisec':
'',
'LastPrice':
'',
'Volume':
0,
'TradingDay':
tradingDay,
'ActionDay':
QA_util_date_str2int(
QA_util_get_last_day(QA_util_date_int2str(tradingDay)))
}
if market_type is None:
market_type = MARKET_TYPE.FUTURE_CN if re.search(
r'[a-zA-z]+', code) else MARKET_TYPE.STOCK_CN
if market_type == MARKET_TYPE.FUTURE_CN:
time_index = time_index_future
else:
time_index = time_index_stock
for item in time_index:
x = str(item).split()[2]
if '.' in x:
tick_pickle['UpdateMillisec'] = 500
else:
tick_pickle['UpdateMillisec'] = 0
tick_pickle['UpdateTime'] = x.split('.')[0]
if (item.seconds >= 75600) & (market_type == MARKET_TYPE.FUTURE_CN):
tick_pickle['ActionDay'] = QA_util_date_str2int(
QA_util_get_last_day(
QA_util_date_int2str(tick_pickle['TradingDay'])))
else:
tick_pickle['ActionDay'] = tick_pickle['TradingDay']
tick_pickle['Volume'] += random.randint(50, 5000)
tick_pickle['LastPrice'] = (ou_noise()+1) * \
weight*price + (1-weight)*price
data.append(copy.deepcopy(tick_pickle))
if ifprint:
print(tick_pickle)
return pd.DataFrame(data)