def __init__(self):
FundHolder.__init__(self)
FundFactor.__init__(self)
FundPool.__init__(self)
FundStatic.__init__(self)
FundExposure.__init__(self)
def cal_weight_date(self, quarter_date):
""" 单个季度公募主动股票基金平均权重 每个基金的权都为1 """
fund_pool = FundPool().get_fund_pool_code(name="普通股票型基金",
date=quarter_date)
for i_fund in range(len(fund_pool)):
fund = fund_pool[i_fund]
try:
asset = self.total_asset.loc[fund, quarter_date] / 100000000
except Exception as e:
asset = 0.5
try:
fund_holding = FundHolder().get_fund_holding_quarter(fund=fund)
fund_holding_date = pd.DataFrame(fund_holding[quarter_date])
fund_holding_date = fund_holding_date.dropna()
fund_holding_date *= asset
fund_holding_date.columns = [fund]
except Exception as e:
fund_holding_date = pd.DataFrame([], columns=[fund])
if i_fund == 0:
stock_data = fund_holding_date
else:
stock_data = pd.concat([stock_data, fund_holding_date], axis=1)
stock_data = stock_data.dropna(how='all')
stock_data_weight = pd.DataFrame(stock_data.sum(axis=1))
return stock_data_weight
def cal_quarter_holding_allfund_quarter(self, quarter_date):
"""
计算 季报日 普通股票+偏股混合基金 基金平均持仓
"""
fund_pool = FundPool().get_fund_pool_code(name="基金持仓基准基金池",
date=quarter_date)
halfyear_date = Date().get_last_fund_halfyear_date(
Date().get_trade_date_offset(quarter_date, 15))
fund_turnover = Fund().get_fund_turnover()
fund_turnover = fund_turnover.loc[fund_pool, :]
fund_turnover[fund_turnover < 15] = np.nan
fund_turnover_date = pd.DataFrame(fund_turnover[halfyear_date])
fund_turnover_date = fund_turnover_date.dropna()
fund_turnover_date = fund_turnover_date.sort_values(by=[halfyear_date],
ascending=True)
fund_pool = list(
fund_turnover_date.index[0:int(len(fund_turnover_date) / 2)])
for i_fund in range(len(fund_pool)):
fund = fund_pool[i_fund]
try:
fund_holding = FundHolder().get_fund_holding_quarter(fund=fund)
fund_holding_date = pd.DataFrame(fund_holding[quarter_date])
fund_holding_date = fund_holding_date.dropna()
fund_holding_date *= 1.0
fund_holding_date.columns = [fund]
except Exception as e:
fund_holding_date = pd.DataFrame([], columns=[fund])
if i_fund == 0:
stock_data = fund_holding_date
else:
stock_data = pd.concat([stock_data, fund_holding_date], axis=1)
stock_data = stock_data.dropna(how='all')
stock_data_weight = pd.DataFrame(stock_data.sum(axis=1))
stock_data_weight.columns = ["WEIGHT"]
stock_data_weight /= stock_data_weight.sum()
stock_data_weight.index.name = "CODE"
sub_path = os.path.join(self.data_weight_path, self.name)
if not os.path.exists(sub_path):
os.makedirs(sub_path)
file = os.path.join(sub_path, quarter_date + '_QuarterHolding.csv')
stock_data_weight.to_csv(file)
def cal_all_wind_file(self):
""" 生成wind文件 """
date_series = Date().get_normal_date_series("20150101",
datetime.today(), "S")
fund_holding = FundHolder().get_fund_holding_halfyear(
fund=self.fund_code)
for i_date in range(len(date_series)):
half_year_date = date_series[i_date]
publish_date = Date().get_trade_date_offset(half_year_date, 0)
try:
fund_holding_date = pd.DataFrame(fund_holding[half_year_date])
fund_holding_date = fund_holding_date.dropna()
fund_holding_date.columns = ["Weight"]
fund_holding_date = fund_holding_date.sort_values(
by=['Weight'], ascending=False)
fund_holding_date["Weight"] /= 100.0
fund_holding_date.loc[
'Cash', 'Weight'] = 1 - fund_holding_date["Weight"].sum()
fund_holding_date.index.name = "Code"
fund_holding_date["CreditTrading"] = "No"
fund_holding_date["Date"] = publish_date
fund_holding_date["Price"] = 0.0
fund_holding_date["Direction"] = "Long"
sub_path = os.path.join(self.wind_port_path, self.port_name)
if not os.path.exists(sub_path):
os.makedirs(sub_path)
file = os.path.join(
sub_path, '%s_%s.csv' % (self.port_name, publish_date))
fund_holding_date.to_csv(file)
except Exception as e:
pass
def cal_fund_holder_exposure_halfyear(self, fund_code, beg_date, end_date):
""" 计算单个基金的半年持仓暴露（注意计算的是非满仓暴露） """
# fund_code, beg_date, end_date = "000001.OF", "20170101", "20190101"
type_list = ['COUNTRY', 'STYLE', 'INDUSTRY']
barra_name = list(Barra().get_factor_name(type_list)['NAME_EN'].values)
out_file = os.path.join(
self.halfyear_exposure_path,
'Fund_Holder_Exposure_HalfYear_%s.csv' % fund_code)
if not os.path.exists(out_file):
beg_date = "20040101"
date_series = Date().get_normal_date_series(beg_date,
end_date,
period='S')
fund_holding = FundHolder().get_fund_stock_weight_halfyear(fund_code)
if fund_holding is not None:
date_series = list(set(date_series) & set(fund_holding.columns))
date_series.sort()
print(date_series)
else:
return None
if len(date_series) > 0:
for i_date in range(0, len(date_series)):
date = date_series[i_date]
report_date = Date().get_normal_date_month_end_day(date)
trade_date = Date().get_trade_date_month_end_day(date)
print("Calculate HalfYear Holder Exposure %s %s" %
(fund_code, report_date))
barra_exposure = Barra().get_factor_exposure_date(
trade_date, type_list)
fund_holding_date = FundHolder(
).get_fund_stock_weight_halfyear(fund_code)
if (barra_exposure is None) or (len(fund_holding_date) == 0):
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
else:
fund_holding_date = pd.DataFrame(fund_holding[report_date])
fund_holding_date = fund_holding_date.dropna()
fund_holding_date = fund_holding_date.sort_values(
by=[report_date], ascending=False)
fund_holding_date.columns = ["Weight"]
fund_holding_date /= 100.0
data = pd.concat([fund_holding_date, barra_exposure],
axis=1)
data = data.dropna()
if (len(data) == 0) or (data is None):
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
else:
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
for i_factor in range(len(barra_name)):
factor_name = barra_name[i_factor]
data_weight = data[['Weight', factor_name]]
data_weight['StockExposure'] = data_weight[
'Weight'] * data_weight[factor_name]
exp = data_weight['StockExposure'].sum()
exposure_add.ix[report_date, factor_name] = exp
country_name = Barra().get_factor_name(
["COUNTRY"])["NAME_EN"].values[0]
position = FundFactor().get_fund_factor(
"Stock_Ratio",
date_list=[report_date],
fund_pool=[fund_code])
exposure_add.ix[
report_date,
country_name] = position.values[0][0] / 100
if i_date == 0:
exposure_new = exposure_add
else:
exposure_new = pd.concat([exposure_new, exposure_add],
axis=0)
else:
exposure_new = pd.DataFrame([])
# 合并新数据
####################################################################
if os.path.exists(out_file):
exposure_old = pd.read_csv(out_file, index_col=[0], encoding='gbk')
exposure_old.index = exposure_old.index.map(str)
params = FactorOperate().pandas_add_row(exposure_old, exposure_new)
else:
params = exposure_new
if len(params) > 0:
params = params[barra_name]
params.to_csv(out_file)
def cal_weight_date(self, date, quarter_date):
days_diff = Date().get_trade_date_diff(quarter_date, date)
fund_pool = FundPool().get_fund_pool_code(name="基金持仓基准基金池",
date=quarter_date)
for i_fund in range(len(fund_pool)):
fund = fund_pool[i_fund]
try:
fund_holding = FundHolder().get_fund_holding_quarter(fund=fund)
fund_holding_date = pd.DataFrame(fund_holding[quarter_date])
fund_holding_date = fund_holding_date.dropna()
fund_holding_date *= 1.0
fund_holding_date.columns = [fund]
except Exception as e:
fund_holding_date = pd.DataFrame([], columns=[fund])
if i_fund == 0:
stock_data = fund_holding_date
else:
stock_data = pd.concat([stock_data, fund_holding_date], axis=1)
stock_data = stock_data.dropna(how='all')
stock_data_weight = pd.DataFrame(stock_data.sum(axis=1))
stock_data_weight /= stock_data_weight.sum()
stock_data_weight.columns = ["Weight"]
stock_data_weight = stock_data_weight.sort_values(by=['Weight'],
ascending=False)
stock_ratio = pd.DataFrame(self.stock_ratio.loc[fund_pool,
quarter_date])
ratio = stock_ratio.median().values[0] / 100.0
if days_diff > 30:
# 得到股票和基金涨跌幅
stock_pool = list(stock_data_weight.index)
beg_date = Date().get_trade_date_offset(date, -61)
date_series = Date().get_trade_date_series(beg_date, date)
f_pct = self.index_return / ratio
s_pct = self.stock_return.loc[date_series, stock_pool]
s_pct = s_pct.T.dropna(how='all').T
s_pct = s_pct.dropna(how='all')
f_pct = f_pct.dropna()
# 准备数据Lasso回归
data = pd.concat([f_pct, s_pct], axis=1)
data = data.loc[beg_date:date, :]
data = data.dropna(subset=['IndexReturn'])
data = data.fillna(0.0)
y = np.row_stack(data['IndexReturn'].values)
x = data.iloc[:, 1:].values
model = LassoCV(fit_intercept=True, positive=True)
# LassoCV自动调节alpha可以实现选择最佳的alpha
model.fit(x, y)
print(model.alpha_)
alpha = model.alpha_
model = Lasso(alpha=alpha, fit_intercept=False, positive=True)
model.fit(x, y)
res = pd.DataFrame(model.coef_[model.coef_ > 0.0001],
index=s_pct.columns[model.coef_ > 0.0001],
columns=[date])
res = res.sort_values(by=[date], ascending=False)
result = pd.concat([res, stock_data_weight], axis=1)
result = result.sort_values(by=['Weight'], ascending=False)
else:
result = stock_data_weight
return result
def cal_fund_holder_exposure_quarter(self, fund, beg_date, end_date):
""" 计算单个基金的季度持仓暴露 (前十大重仓暴露) """
type_list = ['STYLE', 'COUNTRY', 'INDUSTRY']
date_series = Date().get_normal_date_series(beg_date,
end_date,
period='Q')
fund_holding = FundHolder().get_fund_stock_weight_quarter(fund)
if fund_holding is not None:
date_series = list(set(date_series) & set(fund_holding.columns))
date_series.sort()
else:
return None
for i_date in range(0, len(date_series)):
date = date_series[i_date]
report_date = Date().get_normal_date_month_end_day(date)
trade_date = Date().get_trade_date_month_end_day(date)
barra_name = list(
Barra().get_factor_name(type_list)['NAME_EN'].values)
barra_exposure = Barra().get_factor_exposure_date(
trade_date, type_list)
print(
"########## Calculate Quarter Holder Exposure %s %s ##########"
% (fund, report_date))
if (barra_exposure is None) or (fund_holding is None):
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
else:
fund_holding_date = pd.DataFrame(fund_holding[report_date])
fund_holding_date = fund_holding_date.dropna()
fund_holding_date = fund_holding_date.sort_values(
by=[report_date], ascending=False)
fund_holding_date.columns = ["Weight"]
data = pd.concat([fund_holding_date, barra_exposure], axis=1)
data = data.dropna()
if (len(data) == 0) or (data is None):
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
else:
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
for i_factor in range(len(barra_name)):
factor_name = barra_name[i_factor]
data_weight = data[['Weight', factor_name]]
data_weight['StockExposure'] = data['Weight'] * data[
factor_name]
exposure_add.ix[report_date,
factor_name] = data_weight[
'StockExposure'].sum() / 100.0
country_name = Barra().get_factor_name(
["COUNTRY"])["NAME_EN"].values[0]
position = FundFactor().get_fund_factor(
"Stock_Ratio",
date_list=[report_date],
fund_pool=[fund])
position = position.values[0][0]
exposure_add.ix[report_date, country_name] = position / 100
if i_date == 0:
exposure_new = exposure_add
else:
exposure_new = pd.concat([exposure_new, exposure_add], axis=0)
# 合并新数据
####################################################################
out_path = os.path.join(self.data_path_exposure,
'fund_holding_exposure_quarter')
out_file = os.path.join(
out_path, 'Fund_Holder_Exposure_Quarter_' + fund + '.csv')
if os.path.exists(out_file):
exposure_old = pd.read_csv(out_file, index_col=[0], encoding='gbk')
exposure_old.index = exposure_old.index.map(str)
params = FactorOperate().pandas_add_row(exposure_old, exposure_new)
else:
params = exposure_new
params.to_csv(out_file)
def get_data(self):
self.industry_data = FundHolder().get_fund_holding_industry_all()
self.stock_data = FundHolder().get_fund_holding_all()
def cal_fund_holder_exposure(self, fund, beg_date, end_date):
# 每半年计算一次
type_list = ['STYLE', 'COUNTRY', 'INDUSTRY']
date_series = Date().get_normal_date_series(beg_date,
end_date,
period='S')
for i_date in range(len(date_series)):
date = date_series[i_date]
report_date = Date().get_normal_date_month_end_day(date)
trade_date = Date().get_trade_date_month_end_day(date)
barra_name = list(
Barra().get_factor_name(type_list)['NAME_EN'].values)
barra_exposure = Barra().get_factor_exposure_date(
trade_date, type_list)
fund_holding = FundHolder().get_fund_holding_report_date_fund(
fund, report_date)
print("########## Calculate Holder Exposure %s %s ##########" %
(fund, report_date))
if (barra_exposure is None) or (fund_holding is None):
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
else:
fund_holding = fund_holding['Weight']
data = pd.concat([fund_holding, barra_exposure], axis=1)
data = data.dropna()
if (len(data) == 0) or (data is None):
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
else:
exposure_add = pd.DataFrame([],
columns=barra_name,
index=[report_date])
for i_factor in range(len(barra_name)):
factor_name = barra_name[i_factor]
data_weight = data[['Weight', factor_name]]
data_weight['StockExposure'] = data['Weight'] * data[
factor_name]
exposure_add.ix[report_date,
factor_name] = data_weight[
'StockExposure'].sum() / 100.0
if i_date == 0:
exposure_new = exposure_add
else:
exposure_new = pd.concat([exposure_new, exposure_add], axis=0)
# 合并新数据
####################################################################
out_path = Parameter().get_read_file(self.holder_exposure_name)
out_file = os.path.join(out_path,
'Fund_Holder_Exposure_' + fund + '.csv')
if os.path.exists(out_file):
exposure_old = pd.read_csv(out_file, index_col=[0], encoding='gbk')
exposure_old.index = exposure_old.index.map(str)
params = pandas_add_row(exposure_old, exposure_new)
else:
params = exposure_new
params.to_csv(out_file)
def get_fund_holder_data(self, quarter_date, quarter_last_date):
""" 得到数据 """
# date
quarter_trade_date = Date().get_trade_date_offset(quarter_date, 0)
quarter_last_trade_date = Date().get_trade_date_offset(
quarter_last_date, 0)
print(quarter_date, quarter_last_date)
print(quarter_trade_date, quarter_last_trade_date)
# share
data = FundHolder().get_fund_holding_stock_all()
data_quarter = data[data.ReportDate == quarter_date]
data_quarter = data_quarter[data_quarter.PublishDate <= Date().
get_trade_date_offset(quarter_date, 20)]
quarter_share = pd.DataFrame(
data_quarter.groupby(by=['StockCode']).sum()['Share'])
data_quarter = data[data.ReportDate == quarter_last_date]
data_quarter = data_quarter[data_quarter.PublishDate <= Date(
).get_trade_date_offset(quarter_last_date, 20)]
quarter_last_share = pd.DataFrame(
data_quarter.groupby(by=['StockCode']).sum()['Share'])
# price
adjust_factor = Stock().read_factor_h5("AdjustFactor")
price_unadjust = Stock().read_factor_h5("Price_Unadjust")
quarter_adjust = adjust_factor[quarter_trade_date] / adjust_factor[
quarter_last_trade_date]
quarter_price = price_unadjust[quarter_trade_date]
quarter_price_last = price_unadjust[quarter_last_trade_date]
average_price = (
price_unadjust[quarter_trade_date] +
price_unadjust[quarter_last_trade_date] / quarter_adjust) / 2.0
# industry
industry = Stock().read_factor_h5("industry_citic1")
industry_date = pd.DataFrame(industry[industry.columns[-1]])
industry_date.columns = ['Industry']
# concat
result = pd.concat([
quarter_share, quarter_last_share, quarter_price,
quarter_price_last, average_price, quarter_adjust, industry_date
],
axis=1)
result.columns = [
'ShareQuarter', 'ShareQuarterLast', 'PriceQuarter',
'PriceQuarterLast', 'PriceMean', 'Adjust', 'Industry'
]
result = result.dropna(subset=['Adjust', 'Industry'])
result = result.fillna(0.0)
# cal
result['ShareQuarterLastAdjust'] = result['ShareQuarterLast'] * result[
'Adjust']
result['MvQuarter'] = result['ShareQuarter'] * result['PriceQuarter']
result['MvQuarterLast'] = result['ShareQuarterLast'] * result[
'PriceQuarterLast']
result['Inflow'] = (
result['ShareQuarter'] -
result['ShareQuarterLastAdjust']) * result['PriceMean']
result['MvQuarter'] /= 100000000.0
result['Inflow'] /= 100000000.0
result['MvQuarterLast'] /= 100000000.0
return result
def update_data(self):
""" 更新所需要的数据 """
Date().load_trade_date_series("D")
Stock().load_h5_primary_factor()
FundHolder().load_fund_holding_stock()