def _get_data(self):
from single_factor import OperationRevenueGrowth, EstimateNetRevenueGrowRateFY16M, FreeCashFlowPerShare, GrossIncomeRatio, ROC, EffectiveTaxRate, PS
factor_list = [
OperationRevenueGrowth, EstimateNetRevenueGrowRateFY16M,
FreeCashFlowPerShare, GrossIncomeRatio, ROC, EffectiveTaxRate, PS
]
date_one_year = get_trading_date_from_now(self.date, -1, ql.Years)
date_two_year = get_trading_date_from_now(self.date, -2, ql.Years)
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
df_Revenue_growth_one_year = get_factor_from_wind_v2(
self.code_list, [OperationRevenueGrowth], date_one_year)
df_Revenue_growth_one_year.rename(columns={'营业收入增长率': '营业收入增长率_去年'},
inplace=True)
df_FreeCashFlowPerShare_one_year = get_factor_from_wind_v2(
self.code_list, [FreeCashFlowPerShare], date_one_year)
df_FreeCashFlowPerShare_one_year.rename(
columns={'每股企业自由现金流': '每股企业自由现金流_去年'}, inplace=True)
df_FreeCashFlowPerShare_two_year = get_factor_from_wind_v2(
self.code_list, [FreeCashFlowPerShare], date_two_year)
df_FreeCashFlowPerShare_two_year.rename(
columns={'每股企业自由现金流': '每股企业自由现金流_前年'}, inplace=True)
df_temp = df['市销率'] * df['销售毛利率'] * 10.0 # 边际获利乘数即：销售毛利率*10
df = pd.concat([
df, df_Revenue_growth_one_year, df_FreeCashFlowPerShare_one_year,
df_FreeCashFlowPerShare_two_year, df_temp
],
axis=1)
df.rename(columns={0: '高估/低估指数'}, inplace=True)
df = df.dropna()
return df
def _get_data(self):
from single_factor import NetProfitGrowRateV2, NetProfit, Revenue,PE
factor_list = [NetProfitGrowRateV2, NetProfit,PE]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
date_temp = get_trading_date_from_now(self.date, -3, ql.Months)
df_temp = get_factor_from_wind_v2(self.code_list, [NetProfitGrowRateV2, Revenue], date_temp)
df_temp.rename(columns={'净利润增长率': '净利润增长率_3个月前'}, inplace=True)
df_temp_1 = df['净利润增长率'] - df_temp['净利润增长率_3个月前']
df = pd.concat([df, df_temp, df_temp_1], axis=1)
df.rename(columns={0: '差值'}, inplace=True)
df = df.dropna()
return df
def _get_data(self):
from single_factor import PB, PE, DividendYield, PriceFreeCashFlowPerShare, LongTermLiabilityToWorkCapital
factor_list = [PB, DividendYield, PriceFreeCashFlowPerShare, LongTermLiabilityToWorkCapital]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
# 五年PE值获取
df_PE = []
for i in range(5):
date_temp = get_trading_date_from_now(self.date, -i, ql.Years)
df_temp = get_factor_from_wind_v2(self.code_list, [PE], date_temp)
df_temp.rename(columns={'市盈率PE': '市盈率_'+str(i)}, inplace=True)
df_PE.append(df_temp)
df = pd.concat([df]+df_PE, axis=1)
df = df.dropna()
return df
def _get_data(self):
from single_factor import LCap, CurrentRatio, ROE, FreeCashFlowPerShare, OperationRevenueGrowth, NetProfitGrowRateV2
factor_list = [LCap, CurrentRatio, ROE, OperationRevenueGrowth, NetProfitGrowRateV2]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
# 五年自由现金流量
df_PE = []
for i in range(5):
date_temp = get_trading_date_from_now(self.date, -i, ql.Years)
df_temp = get_factor_from_wind_v2(self.code_list, [FreeCashFlowPerShare], date_temp)
df_temp.rename(columns={'每股企业自由现金流指标': '每股企业自由现金流指标_' + str(i)}, inplace=True)
df_PE.append(df_temp)
df = pd.concat([df] + df_PE, axis=1)
df = df.dropna()
return df
def _get_data(self):
from single_factor import DebetToAsset, CFPS, MarketValueToFreeCashFlow, NetProfitGrowRateV2, DividendYield, PE, InventoryTurnRatio
factor_list = [DebetToAsset, CFPS, MarketValueToFreeCashFlow, NetProfitGrowRateV2, DividendYield, PE]
factor_InvTurn_now = [InventoryTurnRatio]
factor_InvTurn_one_year = [InventoryTurnRatio]
date_one_year = get_trading_date_from_now(self.date, -1, ql.Years)
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
# 存货增长率与营收增长率的比较判断数据，使用存货周转率判断
df_invturn_now = get_factor_from_wind_v2(self.code_list, factor_InvTurn_now, self.date)
df_invturn_now.rename(columns={'存货周转率': '存货周转率_今年'}, inplace=True)
df_invturn_one_year = get_factor_from_wind_v2(self.code_list, factor_InvTurn_one_year, date_one_year)
df_invturn_one_year.rename(columns={'存货周转率': '存货周转率_去年'}, inplace=True)
df = pd.concat([df, df_invturn_now, df_invturn_one_year], axis=1)
df = df.dropna()
return df
def output(self):
from single_factor import InventoryTurnRatio
factor_list = [InventoryTurnRatio]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
last_year = self._get_last_year_date(self.date)
df_last_year = get_factor_from_wind_v2(code, factor_list, last_year)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df_last_year.iloc[:, 0] = pd.to_numeric(df_last_year.iloc[:, 0])
growth = (df.iloc[:, 0].values - df_last_year.iloc[:, 0].values
) / df_last_year.iloc[:, 0].values
text = '\0\0存货周转率分析（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '存货周转率为%.2f次；上年同期为%.2f次；同比增长为%.2f%%。'
text = text % (df.iloc[:, 0].values, df_last_year.iloc[:, 0],
growth * 100.0)
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def _get_data(self):
from single_factor import ClosePrice, PE, DividendYield, NetTangibleAssetPerShare, NetLiquidAssetPerShare, TotalLiability, NetLiquidAsset, NetTangibleAsset, CurrentRatio, NetProfitGrowRateV2, PE_MAX
factor_list = [ClosePrice, PE, DividendYield, NetTangibleAssetPerShare, NetLiquidAssetPerShare, TotalLiability, NetLiquidAsset, NetTangibleAsset, CurrentRatio, NetProfitGrowRateV2]
# factor_list = [PE_MAX] # 待添加到上述列表，条件筛选不完整
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
date_temp_1 = get_trading_date_from_now(self.date, -3, ql.Days) # 取3天前国债收益率
yield_data = w.wss("TB" + str(self.N) + "Y.WI", "close", "tradeDate=" + str(date_temp_1) + ";priceAdj=U;cycle=D").Data[0][0]
df = df.dropna()
return df, yield_data
def output(self):
from single_factor import AccRecTurnRatioV2
factor_list = [AccRecTurnRatioV2]
code = [self.code]
date = self._get_last_year_end(self.date)
df = get_factor_from_wind_v2(code, factor_list, date)
date_last = self._get_last_year_date(date)
df_last_year = get_factor_from_wind_v2(code, factor_list, date_last)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df_last_year.iloc[:, 0] = pd.to_numeric(df_last_year.iloc[:, 0])
growth = (df.iloc[:, 0].values - df_last_year.iloc[:, 0].values
) / df_last_year.iloc[:, 0].values
text = '\0\0应收账款率分析（数据截止' + date + '）：'
para_1 = rl_text(text)
text = '应收账款周转率为%.2f次；上年同期为%.2f次；同比增长为%.2f%%。'
text = text % (df.iloc[:, 0].values, df_last_year.iloc[:, 0],
growth * 100.0)
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import AssetTurnoverRatio
factor_list = [AssetTurnoverRatio]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
last_year = self._get_last_year_date(self.date)
df_last_year = get_factor_from_wind_v2(code, factor_list, last_year)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df_last_year.iloc[:, 0] = pd.to_numeric(df_last_year.iloc[:, 0])
growth = (df.iloc[:, 0].values - df_last_year.iloc[:, 0].values
) / df_last_year.iloc[:, 0].values
text = '\0\0管理层面分析_效率分析（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '总资产周转率本年为%.2f次；上年同期为%.2f次；同比增长率为%.2f%%；'
text = text % (df.iloc[:, 0].values, df_last_year.iloc[:, 0],
growth * 100.0)
text_summary = self._get_summary(df, growth)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import ROE
factor_list = [ROE]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
last_year = self._get_last_year_date(self.date)
df_last_year = get_factor_from_wind_v2(code, factor_list, last_year)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df_last_year.iloc[:, 0] = pd.to_numeric(df_last_year.iloc[:, 0])
growth = (df.iloc[:, 0].values - df_last_year.iloc[:, 0].values
) / df_last_year.iloc[:, 0].values
text = '\0\0业绩分析（资本回报率ROE）（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '资本回报为ROE%.2f%%；上年同期为%.2f%%；同比增长率为%.2f%%；'
text = text % (df.iloc[:, 0].values, df_last_year.iloc[:, 0].values,
growth * 100.0)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import NonCurrentAssetRatio
factor_list = [NonCurrentAssetRatio]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
date_last = self._get_last_year_date(self.date)
df_last_year = get_factor_from_wind_v2(code, factor_list, date_last)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df_last_year.iloc[:, 0] = pd.to_numeric(df_last_year.iloc[:, 0])
growth = (df.iloc[:, 0].values - df_last_year.iloc[:, 0].values
) / df_last_year.iloc[:, 0].values
text = '\0\0资产结构分析（数据截止' + date + '）：'
para_1 = rl_text(text)
text = '公司非流动资产占比为%.2f%%；上年同期为%.2f%%；同比增长为%.2f%%；'
text = text % (df.iloc[:, 0].values, df_last_year.iloc[:, 0],
growth * 100.0)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import DebetToAsset, TotalAsset
factor_list = [DebetToAsset, TotalAsset]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
last_year = self._get_last_year_date(self.date)
df_last_year = get_factor_from_wind_v2(code, factor_list, last_year)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df.iloc[:, 1] = pd.to_numeric(df.iloc[:, 1])
df_last_year.iloc[:, 0] = pd.to_numeric(df_last_year.iloc[:, 0])
growth = (df.iloc[:, 0].values - df_last_year.iloc[:, 0].values
) / df_last_year.iloc[:, 0].values
text = '\0\0管理层面分析_效率分析（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '总资产规模为%.2f亿元；资产负债率为%.2f%%；上年同期为%.2f%%；同比增长为%.2f%%；'
text = text % (df.iloc[:, 1].values / 100000000.0,
df.iloc[:, 0].values * 100.0,
df_last_year.iloc[:, 0].values * 100.0, growth * 100.0)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import RevenueGrowthRate, EstimateNetRevenueGrowRateFY16M
factor_list = [RevenueGrowthRate, EstimateNetRevenueGrowRateFY16M]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
df.iloc[:, 1] = pd.to_numeric(df.iloc[:, 1])
text = '\0\0营业收入分析（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '营业收入增长率为%.2f%%；分析师一致预期营业收入增长率为%.2f%%；'
text = text % (df.iloc[:, 0].values, df.iloc[:, 1].values)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import OperateCashFlow, CashFlowCoverRatio
factor_list = [OperateCashFlow, CashFlowCoverRatio]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df.iloc[:, 1] = pd.to_numeric(df.iloc[:, 1])
text = '\0\0现金流分析（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '经营活动现金流为%.2f亿元；' + '利润现金保障倍数为%.2f倍；'
text = text % (df.iloc[:, 0].values / 100000000.0, df.iloc[:,
1].values)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import SellExpense, SellExpenseRevenue, SellExpenseGrowth
factor_list = [SellExpense, SellExpenseRevenue, SellExpenseGrowth]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df.iloc[:, 1] = pd.to_numeric(df.iloc[:, 1])
df.iloc[:, 2] = pd.to_numeric(df.iloc[:, 2])
text = '\0\0经营理念分析——销售费用（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '销售费用为为%.2f亿元；销售费用占营业收入的比重为：%.2f%%；销售费用增长率为%.2f%%；'
text = text % (df.iloc[:, 0].values / 100000000.0, df.iloc[:, 1].values
* 100.0, df.iloc[:, 2].values * 100)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def output(self):
from single_factor import RDExpense, RDExpenseRevenue
factor_list = [RDExpense, RDExpenseRevenue]
code = [self.code]
date = self._get_last_year_end(self.date)
df = get_factor_from_wind_v2(code, factor_list, date)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df.iloc[:, 1] = pd.to_numeric(df.iloc[:, 1])
text = '\0\0经营理念分析——研发费用（数据截止' + date + '）：'
para_1 = rl_text(text)
text = '研发费用为%.2f亿元：占营业收入为%.2f%%。'
text = text % (df.iloc[:, 0].values / 100000000.0,
df.iloc[:, 1].values * 100.0)
para_2 = rl_text(text)
if np.isnan(df.iloc[:, 0].values):
return []
else:
return [para_1, para_2, rl_text('\0')]
def _get_summary(self, df):
from single_factor import RevenueGrowthRate
factor_list = [RevenueGrowthRate]
code = [self.code]
rev_growth = get_factor_from_wind_v2(
code, factor_list, self.date).iloc[:, 0].values / 100.0
sellexp_growth = df.iloc[:, 2]
diff = (rev_growth - sellexp_growth)[0]
if diff >= 0.03:
text_summary = "销售费用的增长慢于销售收入增长，预期公司毛利率具有一定的增长潜力。"
elif diff < -0.03:
text_summary = "销售费用的增长快于销售收入增长，预期公司毛利率增长不乐观。"
elif diff < 0.03 and diff > -0.03:
text_summary = "销售费用的增长与销售收入增长基本持平，预期公司经营稳定。"
elif np.isnan(diff):
text_summary = "公司缺失相关数据，无法分析。"
else:
text_summary = "Warining: recoding is needed!"
return text_summary
def output(self):
from single_factor import NetProfit, NetProfitGrowRateV2, EstimateNetProfitGrowRateFY16M
factor_list = [
NetProfit, NetProfitGrowRateV2, EstimateNetProfitGrowRateFY16M
]
code = [self.code]
df = get_factor_from_wind_v2(code, factor_list, self.date)
df.iloc[:, 0] = pd.to_numeric(df.iloc[:, 0])
df.iloc[:, 1] = pd.to_numeric(df.iloc[:, 1])
df.iloc[:, 2] = pd.to_numeric(df.iloc[:, 2])
text = '\0\0净利润增长分析（数据截止' + self.date + '）：'
para_1 = rl_text(text)
text = '公司净利润为%.2f亿元；净利润增长率为%.2f%%；分析师一致预期净利润增长率为%.2f%%；'
text = text % (df.iloc[:, 0].values / 100000000.0,
df.iloc[:, 1].values, df.iloc[:, 2].values)
text_summary = self._get_summary(df)
text = text + text_summary
para_2 = rl_text(text)
return [para_1, para_2, rl_text('\0')]
def _get_summary(self, df):
from single_factor import NetProfit
factor_list = [NetProfit]
code = [self.code]
netprofit = get_factor_from_wind_v2(code, factor_list,
self.date).iloc[:, 0].values
if df.iloc[:, 0].values < 0.0 and netprofit < 0.0:
text_summary = "此时利润现金保障倍数无实际含义，公司目前整体亏损，公司主营业务经营活动并无现金流入，整体现金流情况较差。"
elif df.iloc[:, 0].values < 0.0 and netprofit > 0.0:
text_summary = "利润现金保障倍数为负值，财务报表中提现公司仍然盈利，但经营活动现金流为负值，说明公司造血能力较差，" \
"日常经营消耗企业现存货币积累。此种情况可以从应收账款、投资活动、筹资活动或者会计记账原则的角度来进一步分析。"
elif df.iloc[:, 0].values > 0.0 and netprofit < 0.0:
text_summary = "企业经营活动现金流为正，但是公司整体亏损，说明公司在主营业务经营活动正常运转，" \
"此类情况并不寻常，可以从成本端和营业外支出/收入中寻找原因。"
elif df.iloc[:, 0].values > 0.0 and netprofit > 0.0:
if df.iloc[:, 1].values > 0.85: # 利润现金保障倍数
text_summary = "净利润和经营现金流都为正数，企业健康发展，运营正常，自身能够产生充足经营现金流，公司产生的净利润是有足够的现金流支持。"
elif df.iloc[:, 1].values <= 0.85 and df.iloc[:, 1].values > 0.5:
text_summary = "净利润和经营现金流都为正数，企业经营活动能够正常运转，但是经营活动现金流入水平较低，公司产生的净利润是有一定的现金流支持，公司经营管理水平有待改善。"
elif df.iloc[:, 1].values <= 0.5 and df.iloc[:, 1].values > 0.0:
text_summary = "净利润和经营现金流都为正数，企业经营活动处于正常运转，但是经营活动现金流入水平偏低，公司产生的净利润缺乏足够的现金流支持，公司经营管理水平有待提高。"
else:
text_summary = "Warining: recoding is needed!"
return text_summary
result_df_list.append(df_industry)
df = pd.concat(result_df_list, axis=0)
return df
class 因子行业去极值(因子去极值):
def get_factor_df(self):
df = self.factor_df
code_list = list(df.index.values) # 股票的代码
factor_list = list(df.columns.values) # 因子名称的代码
SW1_df = SW1Industry(self.date, code_list).get_factor()
df = pd.concat([df, SW1_df], axis=1).dropna() # 去除缺失行业因子的股票
# 根据行业分组进行因子标准化
df_group_by = df.groupby('申万一级行业')
result_df_list = []
for industry, df_industry in df_group_by:
df_industry = df_industry[factor_list]
df_industry = self._deextreme_value(df_industry)
result_df_list.append(df_industry)
df = pd.concat(result_df_list, axis=0)
return df
if __name__ == '__main__':
w.start()
# code_list = w.wset("sectorconstituent", "date=2018-10-30;windcode=000300.SH").Data[1]
code_list = ['000002.SZ', '600000.SH']
factor_df = get_factor_from_wind_v2(code_list, [VOL10, RSI, PE],
'2018-10-30') # 故意引入错误数据
factor_df = 加入行业编码(factor_df, '2018-10-30').get_factor_df()
print(factor_df)
def _get_data(self):
from single_factor import SteadyProfitAcc, InventoryTurnRatio,PB
factor_list = [SteadyProfitAcc, InventoryTurnRatio, PB]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
df = df.dropna()
return df
def _get_data(self):
from single_factor import EstimatePEFY1, EstimateNetProfitGrowRateFY16M, PB
factor_list = [EstimatePEFY1, EstimateNetProfitGrowRateFY16M, PB]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
df = df.dropna()
return df
def _get_data(self):
from single_factor import DilutedEPS, ForecastEarningGrowth_FY1_3M
factor_list = [DilutedEPS, ForecastEarningGrowth_FY1_3M]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
df = df.dropna()
return df
def _get_data(self):
from single_factor import DilutedEPS, EstimateNetProfitGrowRateFY16M
factor_list = [DilutedEPS, EstimateNetProfitGrowRateFY16M]
df = get_factor_from_wind_v2(self.code_list, factor_list, self.date)
df = df.dropna()
return df
