commit model training new example

alphamind/examples/model_training.py

@@ -31,15 +31,16 @@ training     - every 4 week
 
 engine = SqlEngine('postgresql+psycopg2://postgres:A12345678!@10.63.6.220/alpha')
 universe = Universe('zz500', ['zz500'])
-neutralize_risk = ['SIZE'] + industry_styles
-alpha_factors = ['RVOL', 'EPS', 'CFinc1', 'BDTO', 'VAL', 'GREV',
-                 'ROEDiluted']  # ['BDTO', 'RVOL', 'CHV', 'VAL', 'CFinc1'] # risk_styles
+neutralize_risk = industry_styles
+alpha_factors = ['RVOL', 'EPS', 'CFinc1', 'BDTO', 'VAL', 'CHV', 'GREV', 'ROEDiluted']  # ['BDTO', 'RVOL', 'CHV', 'VAL', 'CFinc1'] # risk_styles
 benchmark = 905
 n_bins = 5
 frequency = '1w'
 batch = 4
-start_date = '2011-01-05'
+start_date = '2012-01-01'
 end_date = '2017-08-31'
+method = 'rank'
+use_rank = 100
 
 '''
 fetch data from target data base and do the corresponding data processing
@@ -143,16 +144,23 @@ for i, predict_date in enumerate(dates):
                                               constraints=cons,
                                               detail_analysis=True,
                                               benchmark=benchmark_w,
-                                              is_tradable=is_tradable)
+                                              is_tradable=is_tradable,
+                                              method=method,
+                                              use_rank=use_rank)
+
+    model_res = pd.DataFrame({'weight': model.coef_[0],
+                              'factor': np.array(data_package['x_names'])})
+
+    # model_res.to_csv(r'\\10.63.6.71\sharespace\personal\licheng\portfolio\zz500_model\{0}.csv'.format(predict_date.strftime('%Y-%m-%d')))
 
     final_res[i] = analysis['er']['total'] / benchmark_w.sum()
-    print('trade_date: {0} predicting finished'.format(train_date))
+    print('trade_date: {0} predicting finished'.format(predict_date))
 
 last_date = advanceDateByCalendar('china.sse', dates[-1], frequency)
 
 df = pd.Series(final_res, index=dates[1:] + [last_date])
 df.sort_index(inplace=True)
 df.cumsum().plot()
-plt.title('Risk style factors model training with Linear Regression from 2012 - 2017')
+plt.title('Prod factors model Linear Regression (neutralized Industries rank 100)')
 plt.show()
 

alphamind/model/data_preparing.py

@@ -180,6 +180,7 @@ def fetch_data_package(engine: SqlEngine,
     alpha_logger.info("Data processing is finished")
 
     ret = dict()
+    ret['x_names'] = transformer.names
     ret['settlement'] = return_df
     ret['train'] = {'x': train_x_buckets, 'y': train_y_buckets}
     ret['predict'] = {'x': predict_x_buckets}