update example

alphamind/data/engines/sqlengine.py

@@ -12,6 +12,7 @@ import numpy as np
 import pandas as pd
 import sqlalchemy as sa
 from alphamind.data.engines.universe import Universe
+from PyFin.api import advanceDateByCalendar
 
 risk_styles = ['BETA',
                'MOMENTUM',
@@ -102,9 +103,20 @@ class SqlEngine(object):
 
         return sorted(codes_set)
 
-    def fetch_d1_return(self, ref_date, codes):
+    def fetch_dx_return(self, ref_date, codes, expiry_date=None, horizon=1):
+
+        start_date = ref_date
+
+        if not expiry_date and horizon:
+            end_date = advanceDateByCalendar('china.sse', ref_date, str(horizon) + 'b').strftime('%Y%m%d')
+        elif expiry_date:
+            end_date = expiry_date
+
         codes_str = ','.join(str(c) for c in codes)
-        sql = "select Code, d1 from daily_return where Date = '{ref_date}' and Code in ({codes})".format(ref_date=ref_date,
+        sql = "select Code, sum(d1) as dx from daily_return " \
+              "where Date >= '{start_date}' and Date < '{end_date}'" \
+              " and Code in ({codes}) GROUP BY Code".format(start_date=start_date,
+                                                            end_date=end_date,
                                                             codes=codes_str)
         return pd.read_sql(sql, self.engine)
 

alphamind/examples/example2.py

@@ -9,49 +9,55 @@ import pandas as pd
 from alphamind.analysis.factoranalysis import factor_analysis
 from alphamind.data.engines.sqlengine import SqlEngine
 from alphamind.data.engines.universe import Universe
-from alphamind.data.engines.sqlengine import risk_styles
 from alphamind.data.engines.sqlengine import industry_styles
 from PyFin.api import bizDatesList
 
 engine = SqlEngine('mssql+pymssql://licheng:A12345678!@10.63.6.220/alpha')
 universe = Universe('custom', ['zz500'])
-total_risks = risk_styles + industry_styles
-type = 'risk_neutral'
 
+used_risk_styles = ['SIZE']
 
-def calculate_one_day(ref_date, factors, weights):
+total_risks = used_risk_styles + industry_styles
+build_type = 'risk_neutral'
+
+
+def calculate_one_day(ref_date, factors, factor_weights):
     print(ref_date)
     codes = engine.fetch_codes(ref_date, universe)
     total_data = engine.fetch_data(ref_date, factors, codes, 905)
 
     factor_data = total_data['factor']
-    factor_df = factor_data[['Code', 'industry', 'd1', 'weight', 'isOpen'] + total_risks + factors].dropna()
+    factor_df = factor_data[['Code', 'industry', 'weight', 'isOpen'] + total_risks + factors].dropna()
+
+    dx_return = engine.fetch_dx_return(ref_date, codes)
+    factor_df = pd.merge(factor_df, dx_return, on=['Code'])
 
     weights, _ = factor_analysis(factor_df[factors],
-                                 weights,
+                                 factor_weights,
                                  factor_df.industry.values,
-                                 factor_df.d1.values,
+                                 None,
                                  detail_analysis=False,
                                  benchmark=factor_df.weight.values,
                                  risk_exp=factor_df[total_risks].values,
-                                 is_tradable=factor_df.isOpen.values,
-                                 method=type)
-    return ref_date, (weights.weight - factor_df.weight).dot(factor_df.d1)
+                                 is_tradable=factor_df.isOpen.values.astype(bool),
+                                 method=build_type)
+
+    return ref_date, (weights.weight - factor_df.weight).dot(factor_df.dx)
 
 
 if __name__ == '__main__':
 
     from matplotlib import pyplot as plt
 
-    factors = ['EPS']#, 'FY12P', 'VAL', 'CFinc1', 'BDTO', 'RVOL']
-    weights = [1.]#, 1., 1., 1., 0.5, 0.5]
-    biz_dates = bizDatesList('china.sse', '2013-01-01', '2017-07-05')
+    factors = ['BDTO', 'CFinc1', 'DivP', 'EPS', 'RVOL', 'DROEAfterNonRecurring']
+    factor_weights = [0.10, 0.30, 0.15, 0.18, 0.11, 0.35]
+    biz_dates = bizDatesList('china.sse', '2017-01-01', '2017-07-07')
 
     ers = []
     dates = []
 
     for ref_date in biz_dates:
-        ref_date, er = calculate_one_day(ref_date, factors, weights)
+        ref_date, er = calculate_one_day(ref_date, factors, factor_weights)
         dates.append(ref_date)
         ers.append(er)