def cal_factor_exposure(self, beg_date, end_date):
""" 计算因子暴露 """
# param
long_term = 20
effective_term = int(long_term / 2)
extreme_value = 80
# read data
pct = Stock().read_factor_h5("Pct_chg").T
trade_amount = Stock().read_factor_h5("TradeAmount").T / 100000000
# data precessing
[pct,
trade_amount] = Stock().make_same_index_columns([pct, trade_amount])
trade_amount = trade_amount.fillna(0.0)
# calculate data daily
date_series = Date().get_trade_date_series(beg_date, end_date)
date_series = list(set(date_series) & set(pct.index))
date_series.sort()
res = pd.DataFrame()
for i in range(0, len(date_series)):
current_date = date_series[i]
data_beg_date = Date().get_trade_date_offset(
current_date, -(long_term - 1))
trade_amount_before = trade_amount.loc[
data_beg_date:current_date, :]
if len(trade_amount_before) > effective_term:
print('Calculating factor %s at date %s' %
(self.raw_factor_name, current_date))
zero_number = trade_amount_before.applymap(
lambda x: 1.0 if x == 0.0 else 0.0).sum()
code_filter_list = (
zero_number[zero_number < effective_term]).index
amount_before = trade_amount.loc[data_beg_date:current_date,
code_filter_list]
pct_before = pct.loc[data_beg_date:current_date,
code_filter_list]
iq = pct_before.abs().div(amount_before)
iq[iq > extreme_value] = np.nan
bias = iq.mean()
bias = pd.DataFrame(bias)
bias.columns = [current_date]
else:
print('Calculating factor %s at date %s is null' %
(self.raw_factor_name, current_date))
bias = pd.DataFrame([],
columns=[current_date],
index=trade_amount_before.columns)
res = pd.concat([res, bias], axis=1)
res = res.T.dropna(how='all').T
self.save_alpha_factor_exposure(res, self.raw_factor_name)
def get_balance_data(self, beg_date, end_date):
""" 一段时间内的数据 """
beg_date = Date().get_trade_date_offset(beg_date, 0)
end_date = Date().get_trade_date_offset(end_date, 0)
print(beg_date, end_date)
data = Stock().read_factor_h5("Margin_TradeBalance")
data = data.fillna(0.0)
data /= 100000000.0
concat_data = pd.concat([data[beg_date], data[end_date]], axis=1)
concat_data.columns = ['BegMv', 'EndMv']
concat_data['DiffMv'] = concat_data['EndMv'] - concat_data['BegMv']
concat_data = concat_data.dropna()
return concat_data
def cal_factor_exposure(self, beg_date, end_date):
""" 计算因子暴露 """
# read data
research = Stock().read_factor_h5("ResearchDevelopmentExpense")
research = research.fillna(0.0)
income = Stock().read_factor_h5("OperatingIncome")
research_ttm = Stock().change_single_quarter_to_ttm_quarter(research)
income_ttm = Stock().change_single_quarter_to_ttm_quarter(income) / 4.0
[research_ttm, income_ttm
] = Stock().make_same_index_columns([research_ttm, income_ttm])
roa = research_ttm.div(income_ttm)
report_data = Stock().read_factor_h5("ReportDateDaily")
roa = Stock().change_quarter_to_daily_with_disclosure_date(
roa, report_data, beg_date, end_date)
res = roa.T.dropna(how='all').T
self.save_alpha_factor_exposure(res, self.raw_factor_name)
def VolumeIR(beg_date, end_date):
"""
因子说明：-1* 过去40天成交额标准差 / 过去40天成交额均值
"""
# param
#################################################################################
LongTerm = 40
HalfTerm = int(LongTerm / 2)
factor_name = "VolumeIR"
ipo_num = 90
# read data
#################################################################################
trade_amount = Stock().get_factor_h5("TradeAmount", None, "primary_mfc").T
# code set & date set
#################################################################################
trade_amount = trade_amount.fillna(0.0)
# calculate data daily
#################################################################################
date_series = Date().get_trade_date_series(beg_date, end_date)
date_series = list(set(trade_amount.index) & set(date_series))
date_series.sort()
for i in range(0, len(date_series)):
current_date = date_series[i]
data_beg_date = Date().get_trade_date_offset(current_date,
-(LongTerm - 1))
amount_before = trade_amount.ix[data_beg_date:current_date, :]
if len(amount_before) >= int(0.8 * LongTerm):
print('Calculating factor %s at date %s' %
(factor_name, current_date))
zero_number = amount_before.applymap(lambda x: 1.0
if x == 0.0 else 0.0).sum()
code_filter_list = (zero_number[zero_number < HalfTerm]).index
trade_amount_pre = trade_amount.ix[data_beg_date:current_date,
code_filter_list]
trade_amount_pre_std = trade_amount_pre.std()
trade_amount_pre_mean = trade_amount_pre.mean()
trade_amount_pre_cv = -trade_amount_pre_std / trade_amount_pre_mean
trade_amount_pre_cv = pd.DataFrame(trade_amount_pre_cv.values,
columns=[current_date],
index=trade_amount_pre.columns)
else:
print('Calculating factor %s at date %s is null' %
(factor_name, current_date))
trade_amount_pre_cv = pd.DataFrame([],
columns=[current_date],
index=trade_amount.columns)
if i == 0:
res = trade_amount_pre_cv
else:
res_add = trade_amount_pre_cv
res = pd.concat([res, res_add], axis=1)
res = res.T.dropna(how='all').T
# save data
#############################################################################
Stock().write_factor_h5(res, factor_name, "alpha_dfc")
return res
def VolumeLnMean120d(beg_date, end_date):
"""
因子说明：过去120天的-1*log(交易额)的加权平均 权为随时间线性递减
"""
# param
#################################################################################
LongTerm = 120
HalfTerm = int(LongTerm / 2)
factor_name = 'VolumeLnMean120d'
ipo_num = 90
# read data
#################################################################################
trade_amount = Stock().get_factor_h5("TradeAmount", None, "primary_mfc").T
# code set & date set
#################################################################################
trade_amount = trade_amount.fillna(0.0)
# calculate data daily
#################################################################################
date_series = Date().get_trade_date_series(beg_date, end_date)
date_series = list(set(trade_amount.index) & set(date_series))
date_series.sort()
for i in range(0, len(date_series)):
current_date = date_series[i]
data_beg_date = Date().get_trade_date_offset(current_date,
-(LongTerm - 1))
amount_before = trade_amount.ix[data_beg_date:current_date, :]
if len(amount_before) == LongTerm:
print('Calculating factor %s at date %s' %
(factor_name, current_date))
zero_number = amount_before.applymap(lambda x: 1.0
if x == 0.0 else 0.0).sum()
code_filter_list = (zero_number[zero_number < HalfTerm]).index
amount_before = trade_amount.ix[data_beg_date:current_date,
code_filter_list]
amount_before_log = amount_before.applymap(
lambda x: np.nan if x == 0 else -np.log(x))
weight = np.array(list(range(1, LongTerm + 1)))
weight_amount_log_val = np.dot(amount_before_log.T.values, weight)
weight_amount_log = pd.DataFrame(weight_amount_log_val,
index=amount_before_log.columns,
columns=[current_date])
else:
print('Calculating factor %s at date %s is null' %
(factor_name, current_date))
weight_amount_log = pd.DataFrame([],
columns=[current_date],
index=trade_amount.columns)
if i == 0:
res = weight_amount_log
else:
res_add = weight_amount_log
res = pd.concat([res, res_add], axis=1)
res = res.T.dropna(how='all').T
# save data
#############################################################################
Stock().write_factor_h5(res, factor_name, "alpha_dfc")
return res
def IlliquidityBias(beg_date, end_date):
"""
因子说明： 涨跌幅的绝对值 / 交易额
最近10天均值 / 最近40天均值
"""
# param
#################################################################################
LongTerm = 40
ShortTerm = 10
HalfTerm = LongTerm / 2
factor_name = "IlliquidityBias"
ipo_num = 90
# read data
#################################################################################
pct = Stock().get_factor_h5("Pct_chg", None, "primary_mfc").T
trade_amount = Stock().get_factor_h5("TradeAmount", None, "primary_mfc").T
# data precessing
#################################################################################
[pct, trade_amount] = Stock().make_same_index_columns([pct, trade_amount])
trade_amount = trade_amount.fillna(0.0)
# calculate data daily
#################################################################################
date_series = Date().get_trade_date_series(beg_date, end_date)
date_series = list(set(date_series) & set(pct.index))
date_series.sort()
for i in range(0, len(date_series)):
current_date = date_series[i]
data_beg_date = Date().get_trade_date_offset(current_date,
-(LongTerm - 1))
trade_amount_before = trade_amount.ix[data_beg_date:current_date, :]
if len(trade_amount_before) > HalfTerm:
print('Calculating factor %s at date %s' %
(factor_name, current_date))
zero_number = trade_amount_before.applymap(
lambda x: 1.0 if x == 0.0 else 0.0).sum()
code_filter_list = (zero_number[zero_number < ShortTerm]).index
amount_before = trade_amount.ix[data_beg_date:current_date,
code_filter_list]
pct_before = pct.ix[data_beg_date:current_date, code_filter_list]
illiq = pct_before.abs().div(amount_before,
axis='index') * 100000000
illiq[illiq > 100.0] = np.nan
illiq_bias = illiq.ix[-1 - ShortTerm:, :].mean() / illiq.mean()
price_mean = pd.DataFrame(illiq_bias.values,
columns=[current_date],
index=illiq_bias.index)
else:
print('Calculating factor %s at date %s is null' %
(factor_name, current_date))
price_mean = pd.DataFrame([],
columns=[current_date],
index=trade_amount_before.columns)
if i == 0:
res = price_mean
else:
res = pd.concat([res, price_mean], axis=1)
res = res.T.dropna(how='all').T
# save data
#############################################################################
Stock().write_factor_h5(res, factor_name, "alpha_dfc")
return res
def VolumeUpRatio(beg_date, end_date):
"""
因子说明：1*以当日收盘价为下限 当日收盘价*1.1为上限，
过去120天的在上下限之间的天的成交额的总和占过去120天总成交额的比例
最后乘以 -1
注意：补齐nan为0,，去掉过去120天超过60天交易额为0的股票
"""
# param
#################################################################################
LongTerm = 120
HalfTerm = int(LongTerm / 2)
PctRange = 0.1
factor_name = "VolumeUpRatio"
ipo_num = 90
# read data
#################################################################################
close = Stock().get_factor_h5("PriceCloseAdjust", None, "alpha_dfc").T
trade_amount = Stock().get_factor_h5("TradeAmount", None, "primary_mfc").T
# data precessing
#################################################################################
[close, trade_amount] = Stock().make_same_index_columns([close, trade_amount])
trade_amount = trade_amount.fillna(0.0)
# calculate data daily
#################################################################################
date_series = Date().get_trade_date_series(beg_date, end_date)
date_series = list(set(trade_amount.index) & set(date_series))
date_series.sort()
for i in range(0, len(date_series)):
current_date = date_series[i]
data_beg_date = Date().get_trade_date_offset(current_date, -(LongTerm-1))
amount_before = trade_amount.ix[data_beg_date:current_date, :]
if len(amount_before) >= int(0.8*LongTerm):
print('Calculating factor %s at date %s' % (factor_name, current_date))
zero_number = amount_before.applymap(lambda x: 1.0 if x == 0.0 else 0.0).sum()
code_filter_list = (zero_number[zero_number < HalfTerm]).index
close_low_limit = close.ix[current_date, code_filter_list]
close_up_limit = close.ix[current_date, code_filter_list] * (1 + PctRange)
close_before = close.ix[data_beg_date:current_date, code_filter_list]
price_center = (close_before > close_low_limit) & (close_before < close_up_limit)
trade_amount_filter_sum = amount_before[price_center].sum()
trade_amount_sum = amount_before.sum()
trade_amount_sum = pd.concat([trade_amount_filter_sum, trade_amount_sum], axis=1)
trade_amount_sum.columns = ['filter_sum', 'sum']
trade_amount_sum = trade_amount_sum[trade_amount_sum['sum'] != 0.0]
trade_amount_sum['ratio'] = - trade_amount_sum['filter_sum'] / trade_amount_sum['sum']
else:
print('Calculating factor %s at date %s is null' % (factor_name, current_date))
trade_amount_sum = pd.DataFrame([], columns=['ratio'], index=amount_before.columns)
if i == 0:
res = pd.DataFrame(trade_amount_sum['ratio'].values, columns=[current_date], index=trade_amount_sum.index)
else:
res_add = pd.DataFrame(trade_amount_sum['ratio'].values, columns=[current_date], index=trade_amount_sum.index)
res = pd.concat([res, res_add], axis=1)
res = res.T.dropna(how='all').T
# save data
#############################################################################
Stock().write_factor_h5(res, factor_name, "alpha_dfc")
return res
