-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path20k_Analyzer.py
225 lines (197 loc) · 10.2 KB
/
20k_Analyzer.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
import pandas as pd
import numpy as np
import os
import datetime as dt
import pandas_datareader.data as web
from yahoo_finance import Share
import scipy as sp
from datetime import datetime
import calendar
import sys
# ['underlying_symbol', 'root', 'expiration', 'strike', 'option_type',
# 'open', 'high', 'low', 'close', 'trade_volume', 'bid_size_1545',
# 'bid_1545', 'ask_size_1545', 'ask_1545', 'underlying_bid_1545',
# 'underlying_ask_1545', 'implied_underlying_price_1545',
# 'active_underlying_price_1545', 'implied_volatility_1545', 'delta_1545',
# 'gamma_1545', 'theta_1545', 'vega_1545', 'rho_1545', 'bid_size_eod',
# 'bid_eod', 'ask_size_eod', 'ask_eod', 'underlying_bid_eod',
# 'underlying_ask_eod', 'vwap', 'open_interest', 'delivery_code']
def mt_reporter(data):
report = pd.DataFrame(index=np.arange(1))
report['Expiration'] = str(data.ix[0]['expiration'].date())
report['Day_1_Volume'] = data.ix[0]['trade_volume']
report['Days_to_Expiration'] = ((data.ix[0]['expiration'] - data.index[0])/ np.timedelta64(1, 'D')).astype(int)
report['Option_Symbol'] = ''.join([str(data.ix[0]['expiration'].date()),',',data.ix[0]['option_type'],',',str(data.ix[0]['strike']),',',symb])
report['Trade_Date'] = data.index[0]
tddate = data.index[0]
data = data[1:20]
data['mid_1545'] = (data['bid_1545'] + data['ask_1545'])/2
if len(data) < 19:
ix = pd.DatetimeIndex(start=data.index[-1] + dt.timedelta(days=1), end=data.index[-1] + dt.timedelta(days=19-len(data)), freq='D')
dx = pd.DataFrame(index=ix,columns=data.columns)
data = pd.concat([data,dx],axis=0)
data = data[['open','high','low','close','mid_1545','trade_volume']].replace(to_replace=0, value=np.nan).fillna(method='bfill',axis=1)
if data.ix[0]['open'] != 0:
d2op = data.ix[0]['open']
ds = np.arange(2,len(data)+2)
lolabels = ["D2Lo/D2op", "D3Lo/D2op", "D4Lo/D2op", "D5Lo/D2op", "D6Lo/D2op", "D7Lo/D2op", "D8Lo/D2op", "D9Lo/D2op", "D10Lo/D2op", "D11Lo/D2op", "D12Lo/D2op", "D13Lo/D2op", "D14Lo/D2op", "D15Lo/D2op", "D16Lo/D2op", "D17Lo/D2op", "D18Lo/D2op", "D19Lo/D2op", "D20Lo/D2op"]
hilabels = ["D2Hi/D2op", "D3Hi/D2op", "D4Hi/D2op", "D5Hi/D2op", "D6Hi/D2op", "D7Hi/D2op", "D8Hi/D2op", "D9Hi/D2op", "D10Hi/D2op", "D11Hi/D2op", "D12Hi/D2op", "D13Hi/D2op", "D14Hi/D2op", "D15Hi/D2op", "D16Hi/D2op", "D17Hi/D2op", "D18Hi/D2op", "D19Hi/D2op", "D20Hi/D2op"]
loratios = pd.DataFrame(index=lolabels, columns=np.arange(1))
hiratios = pd.DataFrame(index=hilabels, columns=np.arange(1))
lows = data['low'].reshape(19,1)
highs = data['high'].reshape(19,1)
loratios[[0]] = lows/d2op
hiratios[[0]] = highs/d2op
d2 = data.ix[1]
#d2 = d2[['open','high','low','close','bid_1545','ask_1545']].replace(to_replace=0, value=np.nan).fillna(method='bfill')
report['d2op'] = data.ix[0]['open']
report['d2hi'] = data.ix[0]['high']
report['d2lo'] = data.ix[0]['low']
report['d2cl'] = data.ix[0]['close']
report['d2ivst'] = data.ix[0]['trade_volume']*d2op*10
per1 = data[1:5]
#per1 = per1[['open','high','low','close','bid_1545']].replace(to_replace=0, value=np.nan).fillna(method='bfill',axis=1)
#per1 = per1[['open','high','low','close']].resample('5D').agg({'open': 'first', 'high': 'max','low': 'min', 'close': 'last'}) #need to reformat, giving false values
report['per1op'] = per1.ix[0]['open']
per1op = per1.ix[0]['open']
report['per1hi'] = per1['high'].max()
report['per1lo'] = per1['low'].min()
report['per1cl'] = per1.ix[-1]['close']
report['per1ivst'] = data[2:6]['trade_volume'].mean()*per1op*10
per2 = data[5:10]
#per2 = per2[['open','high','low','close','bid_1545']].replace(to_replace=0, value=np.nan).fillna(method='bfill',axis=1)
#per2 = per2[['open','high','low','close']].resample('4D').agg({'open': 'first', 'high': 'max','low': 'min', 'close': 'last'})#need to reformat, giving false values
report['per2op'] = per2.ix[0]['open']
per2op = per2.ix[0]['open']
report['per2hi'] = per2['high'].max()
report['per2lo'] = per2['low'].min()
report['per2cl'] = per2.ix[-1]['close']
report['per2ivst'] = data[6:11]['trade_volume'].mean()*per2op*10
try:
per3 = data.ix[10:]
#per3 = per3[['open','high','low','close','bid_1545']].replace(to_replace=0, value=np.nan).fillna(method='bfill',axis=1)
#per3 = per3[['open','high','low','close']].resample('11D').agg({'open': 'first', 'high': 'max','low': 'min', 'close': 'last'}) #need to reformat, giving false values
report['per3op'] = per3.ix[0]['open']
per3op = per3.ix[0]['open']
report['per3hi'] = per3['high'].max()
report['per3lo'] = per3['low'].min()
report['per3cl'] = per3.ix[-1]['close']
report['per3ivst'] = data[10:]['trade_volume'].mean()*per3op*10
report['Avg_Investible_Volume'] = (report['per3ivst']+report['per2ivst']+report['per1ivst']+report['d2ivst'])/4
except:
report['per3op'] = 'NA'
per3op = 'NA'
report['per3hi'] = 'NA'
report['per3lo'] = 'NA'
report['per3cl'] = 'NA'
report['per3ivst'] = 0
report['Avg_Investible_Volume'] = (report['per2ivst']+report['per1ivst']+report['d2ivst'])/3
report['Trade_Hi'] = hiratios.max()
report['Trade_Lo'] = loratios.min()
report['per1hi_percent'] = report['per1hi']/d2op
report['per2hi_percent'] = report['per2hi']/d2op
report['per3hi_percent'] = report['per3hi']/d2op
report['per1lo_percent'] = report['per1lo']/d2op
report['per2lo_percent'] = report['per2lo']/d2op
report['per3lo_percent'] = report['per3lo']/d2op
report['per1cl_percent'] = report['per1cl']/d2op
report['per2cl_percent'] = report['per2cl']/d2op
report['per3cl_percent'] = report['per3cl']/d2op
loratios = loratios.T
hiratios = hiratios.T
report['Trade_Hi_Profit'] = report['Trade_Hi'] -1
rprt = pd.concat([report,hiratios,loratios],axis=1)
else:
rprt = np.nan
return rprt
def next_monthly(d): # must test len of option price history to ensure 1: that purchase is possible on d2 and 2: that there is at least 12 days of data
f = d + dt.timedelta(days=30)
f = (f + dt.timedelta(days=(calendar.FRIDAY - f.weekday()) % 7))
e = d + dt.timedelta(days=(calendar.FRIDAY - f.weekday()) % 7)
return e,f
def find_strike(array,value,movement):
if len(array) > 1:
array = np.sort(array, axis=0, kind='quicksort', order=None)
idx = np.abs(array-value).argmin()
stx = array[idx+movement]
else:
stx = array
return stx
def 20k_sortby(data,volume,days_to_expiration):
underlying = pd.Series(data=((data['underlying_bid_eod']+data['underlying_ask_eod'])/2),index=data.index)
stx = pd.Series(data=(data['strike'] -underlying),index=data.index)
d2x = pd.Series(data=(data['expiration']-data['quote_date']/np.timedelta64(1, 'D')).astype(int), index=data.index)
sorted_data = data[(data['trade_volume']>=volume)&(np.in1d(d2x,days_to_expiration))&(np.abs(stx)<underlying*1.05)]
return sorted_data
def simple_reporter(data):
report = pd.DataFrame(index=np.arange(1))
report['Expiration'] = str(data.ix[0]['expiration'].date())
report['Day_1_Volume'] = data.ix[0]['trade_volume']
report['Days_to_Expiration'] = ((data.ix[0]['expiration'] - data.index[0])/ np.timedelta64(1, 'D')).astype(int)
report['Option_Symbol'] = ''.join([str(data.ix[0]['expiration'].date()),',',data.ix[0]['option_type'],',',str(data.ix[0]['strike']),',',symb])
report['Trade_Date'] = data.index[0]
d2op = data.ix[1]['open']
report['d2op'] = data.ix[1]['open']
report['d2hi'] = data.ix[1]['high']
report['d2lo'] = data.ix[1]['low']
report['d2cl'] = data.ix[1]['close']
report['d2ivst'] = data.ix[1]['trade_volume']*d2op*10
report['Trade_Hi/d2op'] = data.ix[1:]['high'].max()/d2op
report['Trade_Lo/d2op'] = data.ix[1:]['high'].max()/d2op
return report
def spike_finder(data,volume,days_to_expiration,col,q):
underlying = pd.Series(data=((data['underlying_bid_eod']+data['underlying_ask_eod'])/2),index=data.index)
stx = pd.Series(data=(data['strike'] -underlying),index=data.index)
d2x = pd.Series(data=(data['expiration']-data['quote_date']/np.timedelta64(1, 'D')).astype(int), index=data.index)
sorted_data = data[(data['trade_volume']>=volume)&(np.in1d(d2x,days_to_expiration))&(np.abs(stx)<underlying*1.05)]
spike_level = np.abs(sorted_data[col].quantile(q))
return spike_level
types = ('put','call')
symbs = ('AAPL','')
years = ('2016','2017')
c = 'trade_volume'
relation = '>='
value = 20000
for symb in symbs:
dir = 'C:\\Users\\asus\\Documents\\Quant\\Database\\' + symb + '\\Options\\'
os.makedirs('C:\\Users\\asus\\Dropbox\\Outlines\\MTAUTO-PYTHON\\Sort_By\\', exist_ok=True)
rprtsortbypath = 'C:\\Users\\asus\\Dropbox\\Outlines\\MTAUTO-PYTHON\\Sort_By\\' 'Sort_By_Summary_Report_' + symb + '_' + c + '-' + str(value) + '.csv'
rawsortbypath = 'C:\\Users\\asus\\Dropbox\\Outlines\\MTAUTO-PYTHON\\Sort_By\\' 'Sort_By_Raw_Data_' + symb + '_' + c + str(value) +'.csv'
report = pd.DataFrame()
rawreport = pd.DataFrame()
for type in types:
if symb == '^VIX':
symb = 'VXX'
for year in years:
new_year = pd.to_datetime(year+'-01-01', infer_datetime_format=True)
if type == 'call':
diropt = ''.join([dir,symb,'_',year,'_Calls.csv'])
x = 1
elif type == 'put':
diropt = ''.join([dir,symb,'_',year,'_Puts.csv'])
x = -1
try:
data = pd.read_csv(diropt,'rb',delimiter=',',parse_dates=['expiration','quote_date'],infer_datetime_format=True).sort_index(axis=0)
except:
continue
#filter near the money options
data.drop_duplicates(['expiration', 'strike','quote_date'],inplace=True)
sorted_data = 20k_sortby(data=data,volume=value,days_to_expiration=[5,4])
i = 0
for i in sorted_data.index:
opt = data[(data['expiration'] == sort.ix[i]['expiration'])&(data['strike'] == sort.ix[i]['strike'])].set_index('quote_date')
opt = opt[sorted_data.ix[i]['quote_date']:]
if (len(opt) > 2):
if (opt.ix[1]['open'] != 0):
rprt = simple_reporter(opt)
i = i + 1
rprt['Trade_#'] = i
opt['Trade_#'] = i
report = pd.concat([report,rprt],axis=0)
rawreport = pd.concat([rawreport,opt],axis=0)
else:
continue
else:
continue
report.to_csv(rprtsortbypath)
rawreport.to_csv(rawsortbypath)