def main(stocks=None, args=[1, 2, 3, 4, 5]):
if stocks == None:
stocks = tools.get_stocks()
data = {
stock:
pd.read_csv('../../DataBase/StockDailyData/Stock/%s.csv' % stock,
index_col=[0],
parse_dates=[0])
for stock in stocks
}
OPEN = DataFrame({stock: data[stock].loc[:, 'open'] for stock in stocks})
HIGH = DataFrame({stock: data[stock].loc[:, 'high'] for stock in stocks})
LOW = DataFrame({stock: data[stock].loc[:, 'low'] for stock in stocks})
CLOSE = DataFrame({stock: data[stock].loc[:, 'close'] for stock in stocks})
ADJ = DataFrame(
{stock: data[stock].loc[:, 'adj_factor']
for stock in stocks})
st = DataFrame({stock: data[stock].loc[:, 'st'] for stock in stocks})
AMOUNT = DataFrame(
{stock: data[stock].loc[:, 'amount']
for stock in stocks})
st = st.shift()
no_liquid = (AMOUNT.lt(AMOUNT.rolling(5).mean().quantile(0.05, axis=1),
axis=0)).shift()
tingpai = (CLOSE == np.nan) | (AMOUNT == 0)
CLOSE = (np.log(CLOSE * ADJ)).fillna(method='ffill')
OPEN = (np.log(OPEN * ADJ)).fillna(value=CLOSE)
HIGH = (np.log(HIGH * ADJ)).fillna(value=CLOSE)
LOW = (np.log(LOW * ADJ)).fillna(value=CLOSE)
yiziban = (HIGH == LOW) & (HIGH > CLOSE.shift())
y1 = OPEN.shift(-2) - OPEN.shift(-1)
y2 = OPEN.shift(-3) - OPEN.shift(-2)
y3 = OPEN.shift(-4) - OPEN.shift(-3)
y4 = OPEN.shift(-5) - OPEN.shift(-4)
y5 = OPEN.shift(-6) - OPEN.shift(-5)
def list_n_na(s, n):
for i in range(n):
s.loc[s.first_valid_index()] = np.nan
return s
y1 = y1.apply(func=list_n_na, args=(60, ), axis=0, result_type='expand')
y2 = y2.apply(func=list_n_na, args=(60, ), axis=0, result_type='expand')
y3 = y3.apply(func=list_n_na, args=(60, ), axis=0, result_type='expand')
y4 = y4.apply(func=list_n_na, args=(60, ), axis=0, result_type='expand')
y5 = y5.apply(func=list_n_na, args=(60, ), axis=0, result_type='expand')
y1[st | no_liquid | yiziban | tingpai] = np.nan
y2[st | no_liquid | yiziban | tingpai] = np.nan
y3[st | no_liquid | yiziban | tingpai] = np.nan
y4[st | no_liquid | yiziban | tingpai] = np.nan
y5[st | no_liquid | yiziban | tingpai] = np.nan
y1.to_csv('../Data/y1.csv')
y2.to_csv('../Data/y2.csv')
y3.to_csv('../Data/y3.csv')
y4.to_csv('../Data/y4.csv')
y5.to_csv('../Data/y5.csv')
})
ADJ = DataFrame({
stock: pd.read_csv('%s/StockDailyData/Stock/%s.csv' %
(gc.DATABASE_PATH, stock),
index_col=[0],
parse_dates=[0]).loc[:, 'adj_factor']
for stock in self.stocks
})
CLOSE = CLOSE * ADJ
r = np.log(CLOSE).diff()
n = 20
a = r.rolling(n).mean() / r.rolling(n).std()
a = a.loc[a.index >= self.start_date, :]
a = a.loc[a.index <= self.end_date, :]
self.factor = a
#%%
if __name__ == '__main__':
#获取股票
stocks = tools.get_stocks()
a = Sharpe('Sharpe',
stocks=stocks,
start_date='20200101',
end_date='20201010')
a.generate_factor()
a.factor_analysis()
def main(start_date, end_date):
#获取股票
stocks = tools.get_stocks()
#获取行业
industrys = tools.get_industrys(level='L1', stocks=stocks)
industrys = {k: industrys[k] for k in industrys.keys()}
stocks = []
for v in industrys.values():
stocks.extend(v)
stocks.sort()
CLOSE = DataFrame({
stock: pd.read_csv('%s/StockDailyData/Stock/%s.csv' %
(gc.DATABASE_PATH, stock),
index_col=[0],
parse_dates=[0]).loc[:, 'close']
for stock in stocks
})
dates = CLOSE.index
for ind in industrys.keys():
if len(industrys[ind]) > 0:
df = DataFrame(0, index=dates, columns=stocks)
df.loc[:, industrys[ind]] = 1
if os.path.exists('%s/Data/%s.csv' % (gc.FACTORBASE_PATH, ind)):
df_old = pd.read_csv('%s/Data/%s.csv' %
(gc.FACTORBASE_PATH, ind),
index_col=[0],
parse_dates=[0])
df = pd.concat([df_old, df.loc[df.index > df_old.index[-1]]],
axis=0)
df.sort_index(0, inplace=True)
df.sort_index(1, inplace=True)
df.to_csv('%s/Data/%s.csv' % (gc.FACTORBASE_PATH, ind))
#遍历取pickle
files = os.listdir('./')
files = list(filter(lambda x: len(x) > 4, files))
factors_1 = list(filter(lambda x: x[-5:] == '_1.py', files))
factors_2 = list(filter(lambda x: x[-5:] == '_2.py', files))
#生成单因子
for p in factors_1:
if os.path.exists('%s/Base/%s.csv' %
(gc.FACTORBASE_PATH, p.split('.')[0][:-2])):
start_date = datetime.datetime.today().strftime('%Y%m%d')
end_date = datetime.datetime.today().strftime('%Y%m%d')
else:
start_date = '20200101'
end_date = datetime.datetime.today().strftime('%Y%m%d')
exec('from %s import %s' % (p.split('.')[0], p.split('.')[0][:-2]))
factor = eval('%s("%s", stocks, start_date, end_date)' %
(p.split('.')[0][:-2], p.split('.')[0][:-2]))
factor.update_factor()
#生成合成因子
for p in factors_2:
if os.path.exists('%s/Base/%s.csv' %
(gc.FACTORBASE_PATH, p.split('.')[0][:-2])):
start_date = datetime.datetime.today().strftime('%Y%m%d')
end_date = datetime.datetime.today().strftime('%Y%m%d')
else:
start_date = '20200101'
end_date = datetime.datetime.today().strftime('%Y%m%d')
exec('from %s import %s' % (p.split('.')[0], p.split('.')[0][:-2]))
factor = eval('%s("%s", stocks, start_date, end_date)' %
(p.split('.')[0][:-2], p.split('.')[0][:-2]))
factor.update_factor()