In [51]:

    

# System libraries
import sys, os, gc
import datetime
import itertools

# Math libraries
import math, random
import pandas as pd, numpy as np
import scipy
from scipy import stats
import datetime
from datetime import timedelta
#from datetime import datetime

# Data storage libraries
import pickle, sqlite3, simpledbf, boto3

# Custom financial data libraries
import utils.findata_utils as fd

# Plotting libraries
import matplotlib.pyplot as plt
from matplotlib import rcParams

db = 'C:\\Datasets\\thesis.db'
conn = sqlite3.connect(db)
c = conn.cursor()

overleaf = ['C:','Users','bryce','OneDrive','Documents','Overleaf','Thesis','assets','exports']


    

In [9]:

    

returns = pd.read_sql('''SELECT * FROM daily_abnormal_returns''', conn)


    

In [12]:

    

returns['d'] = returns.apply(lambda r: (datetime.datetime.strptime(r['date'],'%Y-%m-%d') - 
                                        datetime.datetime.strptime(str(r['rebal_date'])[:10],'%Y-%m-%d')).days,
                             axis=1)


    

In [16]:

    

plt.style.use('classic')

fig = plt.figure(figsize=(10,5))
fig.patch.set_facecolor('white')

ax = fig.add_subplot(1, 1, 1)

ax.spines['left'].set_visible(True)
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
ax.spines['bottom'].set_visible(True)
ax.grid(True,axis='both',linestyle=':')


agg = returns

ax.plot(returns.loc[returns['d'].isbetween(-30,30)]['d'],
        returns.loc[(returns['d']<=30) & (returns['d']>=-30)]['ar_daily'].mean(),
                                          label='Abnormal returns')



ax.legend(frameon=False, loc='best')

plt.title('Portfolio returns')
plt.ylabel('Culmulative returns')
plt.xlabel('Days Since (To) Rebalance')

xlim = [-365,365]

plt.xlim(time_range)
plt.ylim(-0.5,0.5)
plt.show()


    

    




---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-16-f96272815920> in <module>
     15 
     16 
---> 17 ax.plot(returns.loc[returns['d'].isbetween(-30,30)]['d'],
     18         returns.loc[(returns['d']<=30) & (returns['d']>=-30)]['ar_daily'].mean(),
     19                                           label='Abnormal returns')

C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\generic.py in __getattr__(self, name)
   5055             if self._info_axis._can_hold_identifiers_and_holds_name(name):
   5056                 return self[name]
-> 5057             return object.__getattribute__(self, name)
   5058 
   5059     def __setattr__(self, name, value):

AttributeError: 'Series' object has no attribute 'isbetween'





    

In [20]:

    

model_inputs = pd.read_sql('''SELECT * FROM model_inputs''',conn)

display(model_inputs.head())


    

    








  

    

      

      
pct_contrib
      
turnover
      
VWAP
      
vol
      
VWMC
      
SPTSXComp
      
add
    
  
  

    

      
0
      
0.000080
      
0.047819
      
0.547673
      
3008101.0
      
5.260527e+07
      
6.545272e+11
      
0
    
    

      
1
      
0.000091
      
0.035671
      
0.632663
      
2243882.0
      
6.076876e+07
      
6.677543e+11
      
0
    
    

      
2
      
0.000108
      
0.031565
      
0.780580
      
1985592.0
      
7.497652e+07
      
6.935705e+11
      
0
    
    

      
3
      
0.000099
      
0.039976
      
0.715963
      
2514749.0
      
6.876985e+07
      
6.950506e+11
      
0
    
    

      
4
      
0.000128
      
0.050814
      
0.968178
      
3196490.0
      
9.299567e+07
      
7.271307e+11
      
0
    
  

In [52]:

    

for factor in model_inputs.columns:
    fig = plt.figure(figsize=(6,6))
    fig.patch.set_facecolor('white')
    ax = fig.add_subplot(1, 1, 1)
    ax.spines['left'].set_visible(True)
    ax.spines['top'].set_visible(False)
    ax.spines['right'].set_visible(False)
    ax.spines['bottom'].set_visible(True)
    ax.grid(True,axis='both',linestyle=':')
    
    ax.hist(model_inputs.loc[model_inputs['add']==1][factor], density=True, bins=np.linspace(0,0.02,10), alpha=0.7, cumulative=False, histtype='stepfilled', label='Add', color=['xkcd:navy blue'])
    ax.hist(model_inputs.loc[model_inputs['add']==0][factor], density=True, bins=np.linspace(0,0.02,10), alpha=0.4, cumulative=False, histtype='stepfilled', label='n/c', color=['xkcd:forest green'])
    #plt.hist(y_train)
    plt.title('Distribution of {factor}'.format(factor=factor))
    plt.ylabel('Frequency')
    plt.xlabel('{factor}'.format(factor=factor))
    plt.legend(frameon=False, loc='top right')
    
    plt.xlim(0,0.02)
    
    #plt.xticks(np.linspace(0,0.02,5), np.linspace(0,2,5))
    plt.show()
    
    fig.savefig('\\'.join(overleaf+['distribution_{factor}.png'.format(factor=factor)]))


    

    




C:\ProgramData\Anaconda3\lib\site-packages\matplotlib\legend.py:497: UserWarning: Unrecognized location "top right". Falling back on "best"; valid locations are
	best
	upper right
	upper left
	lower left
	lower right
	right
	center left
	center right
	lower center
	upper center
	center

  % (loc, '\n\t'.join(self.codes)))






    













    













    




C:\ProgramData\Anaconda3\lib\site-packages\numpy\lib\histograms.py:823: RuntimeWarning: invalid value encountered in true_divide
  return n/db/n.sum(), bin_edges