-
Notifications
You must be signed in to change notification settings - Fork 17
/
Copy pathclean_data.py
187 lines (160 loc) · 6.89 KB
/
clean_data.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
import numpy as np
import pandas as pd
import h5py
import matplotlib.pyplot as plt
def load_messages(path, name, date):
data = h5py.File(path, 'r')
messages = data['/messages/' + name + '/' + date]
mdata = messages[:, :]
t, n = mdata.shape
data.close()
mcolumns = ['msec',
'type',
'buysell',
'price',
'shares',
'refno']
mout = pd.DataFrame(mdata, index=np.arange(0, t), columns=mcolumns)
# mout["time"]= pd.to_datetime(mout["msec"],unit='ms',origin=pd.Timestamp(date))
# mout = mout[['time','msec', 'type', 'buysell', 'price', 'shares', 'refno']]
return mout
def load_books(path, name, date):
data = h5py.File(path, 'r')
orderbooks = data['/orderbooks/' + name + '/' + date]
mdata = orderbooks[:, :]
t, n = mdata.shape
data.close()
# columns names
time = ['msec']
bid_price = ['bp'+str(i) for i in range(1,11)]
ask_price = ['ap'+str(i) for i in range(1,11)]
bid_volumn = ['bv'+str(i) for i in range(1,11)]
ask_volumn = ['av'+str(i) for i in range(1,11)]
mcolumns = time+bid_price+ask_price+bid_volumn+ask_volumn
mout = pd.DataFrame(mdata, index=np.arange(0, t),columns=mcolumns)
return mout
def vwap_series(df, tinterval):
df['sec'] = df['msec']/1000
vwap_list = []
df_v = df.values
time = 34200
temp = []
for i in range(len(df_v)):
if df_v[i][6]>= time and df_v[i][6]< time+tinterval:
temp.append([df_v[i][3],df_v[i][4]])
if df_v[i][6]>= time+tinterval:
time = time+tinterval
vol_time_price = [x[0]*x[1] for x in temp]
if sum([x[1] for x in temp]) != 0:
vwap_list.append(sum(vol_time_price)/sum([x[1] for x in temp]))
temp = []
else:
vwap_list.append(np.nan)
temp = []
if i == len(df_v)-1:
# multiply volume by price for each row in the 10s interval
vol_time_price = [x[0]*x[1] for x in temp]
if sum([x[1] for x in temp]) != 0:
# sum all the vol*p and divide by total volume to get vwap
vwap_list.append(sum(vol_time_price)/sum([x[1] for x in temp]))
else:
vwap_list.append(np.nan)
return vwap_list
DATE_list = ['20181105','20181106','20181107','20181108','20181109',
'20181112','20181113','20181114','20181115','20181116',
'20181119','20181120','20181121','20181126',
'20181127','20181128','20181129','20181130','20181203',
'20181204' ]
df_mult_date = pd.DataFrame()
for DATE in DATE_list:
# Goldman Sachs message data
# df = load_messages('gs_tick_data.hdf5', 'GS', DATE)
# Apple message data
df = load_messages('/Volumes/easystore/FML_project/aapl_tick_data.hdf5', 'AAPL', DATE)
df = df[(df['msec'] >= 34200000) & (df['msec'] <= 57600000)]
ex = df[df['type'].isin([2,4,6,7])]
ex = ex.reset_index(drop=True)
ex['price'] = ex['price']/10000
# computing 10 second vwap series
vwap = vwap_series(ex,10)
sec = list(range(34210,57601,10))
vwap_df = pd.DataFrame()
vwap_df['vwap'] = vwap
vwap_df['sec'] = sec
vwap_df['msec'] = vwap_df['sec']*1000
vwap_df["time"]= pd.to_datetime(vwap_df["msec"],unit='ms',origin=pd.Timestamp(20181204))
vwap_df = vwap_df.dropna()
# l.plot(x='time', y='vwap')
# plt.show()
# Goldman Sachs Tick data
# book = load_books('gs_tick_data.hdf5', 'GS', DATE)
# Apple tick data
book = load_books('/Volumes/easystore/FML_project/aapl_tick_data.hdf5','AAPL',DATE)
book = book[(book['msec'] >= 34200000) & (book['msec'] <= 57600000)]
book = book.reset_index(drop=True)
book = book[['msec', 'bp1', 'bp2', 'bp3', 'bp4', 'bp5', 'ap1', 'ap2', 'ap3', 'ap4', 'ap5', 'bv1', 'bv2', 'bv3', 'bv4', 'bv5', 'av1', 'av2', 'av3', 'av4', 'av5']]
book['origion'] = 1
msec = [x*1000 for x in list(range(34200,57600,10))]
mcolumns = ['msec', 'bp1', 'bp2', 'bp3', 'bp4', 'bp5', 'ap1', 'ap2', 'ap3', 'ap4', 'ap5', 'bv1', 'bv2', 'bv3', 'bv4', 'bv5', 'av1', 'av2', 'av3', 'av4', 'av5','origion']
a = np.empty((len(msec),22,))
a[:] = np.nan
insert_book = pd.DataFrame(a,index=np.arange(0, len(msec)),columns=mcolumns)
insert_book['msec'] = msec
insert_book['origion'] = 0
# merge
frames = [book, insert_book]
y = pd.concat(frames,ignore_index=True)
y = y.sort_values(by=['msec'])
y = y.fillna(method='ffill')
# pull out
new_book = y[y['origion']==0]
new_book = new_book.dropna()
new_book = new_book.drop(columns=['origion'])
d = vwap_df.join(new_book.set_index('msec'), on='msec')
d = d.dropna()
d = d.drop(columns=['time','sec'])
cols = ['bp1', 'bp2', 'bp3', 'bp4', 'bp5', 'ap1', 'ap2', 'ap3', 'ap4', 'ap5', 'bv1', 'bv2', 'bv3', 'bv4', 'bv5', 'av1', 'av2', 'av3', 'av4', 'av5']
for col in cols:
d['delta_'+col] = d[col].diff(1)
d['mean_volumn_diff'] = (d['bv1']+d['bv2']+d['bv3']+d['bv4']+d['bv5'])/5 - (d['av1']+d['av2']+d['av3']+d['av4']+d['av5'])/5
d['spread'] = d['ap1'] - d['bp1']
d['vol_unb1'] = (d['bv1'] - d['av1'])/d['bv1']
d['vol_unb2'] = (d['bv2'] - d['av2'])/d['bv2']
d['vol_unb3'] = (d['bv3'] - d['av3'])/d['bv3']
d['vol_unb4'] = (d['bv4'] - d['av4'])/d['bv4']
d['vol_unb5'] = (d['bv5'] - d['av5'])/d['bv5']
d_v = d.values
mom_b = [np.nan,np.nan,np.nan,np.nan,np.nan]
volat_b = [np.nan,np.nan,np.nan,np.nan,np.nan]
mom_a = [np.nan,np.nan,np.nan,np.nan,np.nan]
volat_a = [np.nan,np.nan,np.nan,np.nan,np.nan]
# why volataility for last 5
for i in range(5,len(d_v)):
bp_past5 = np.asarray([d_v[i-1][2]/10000,d_v[i-2][2]/10000,d_v[-3][2]/10000,d_v[-4][2]/10000,d_v[i-5][2]/10000])
ap_past5 = np.asarray([d_v[i-1][7]/10000,d_v[i-2][7]/10000,d_v[-3][7]/10000,d_v[-4][7]/10000,d_v[i-5][7]/10000])
mom_b.append((d_v[i][2]-d_v[i-5][2])/d_v[i-5][2])
volat_b.append(bp_past5.std())
mom_a.append((d_v[i][7]-d_v[i-5][7])/d_v[i-5][7])
volat_a.append(ap_past5.std())
d['mom_bp1'] = mom_b
d['mom_ap1'] = mom_a
d['vola_bp1'] = volat_b
d['vola_ap1'] = volat_a
label1 = []
label2 = []
for i in range(len(d_v)-1):
if d_v[i+1][0]>d_v[i][0]:
label1.append(1)
if d_v[i+1][0]<d_v[i][0]:
label1.append(-1)
label2.append(d_v[i+1][0])
label1.append(np.nan)
label2.append(np.nan)
d['vwap_d'] = label1
d['vwap_v'] = label2
d = d.dropna()
d = d.reset_index(drop = True)
frames = [df_mult_date, d]
df_mult_date = pd.concat(frames,ignore_index=True)
# AAPL limit order features data
df_mult_date.to_csv('labelled_data_10s_AAPL', index=False)