-
Notifications
You must be signed in to change notification settings - Fork 0
/
hft_signal_analysis.py
179 lines (138 loc) · 5.37 KB
/
hft_signal_analysis.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
import hft_signal_lib
reload(hft_signal_lib)
from hft_signal_lib import *
import numpy as np
import datetime
import glob
import os
from collections import defaultdict
### some code to analysis index futures
if len(sys.argv) <= 1:
yearmonth = '201101'
else :
yearmonth = sys.argv[1]
result = load_data(IFHV, yearmonth)
dates = result.keys()
dates.sort()
decays = [0.01, 0.1, 0.5, 0.7, 0.8, 0.9, 0.95, 0.975, 0.99, 0.995, 0.999]
decay_names = [str(ff).replace('0.', '_') for ff in decays]
burn_in = 60
offsets = range(2, 60, 4)
#data = result[dates[0]]
### signals
def get_signals(data):
mid = midpoint(data)
book = booksignal(data, 1, 0.9)
last = data['lastprice']
emabook = [ema(book, ff) for ff in decays]
vwap = [vwap_ema(data, ff) for ff in decays]
trends = [book / ema(book, ff) for ff in decays]
trsigns = [tradesignema(data, ff) for ff in decays]
trsvolumes = [tradesignvolume_ema(data, ff) for ff in decays]
sv = [signedvolume_ema(data, ff) for ff in decays]
spr = spread(data)
base_ind = subsample_nsample(data, 60)
base_ind, offsets_ind = generate_offset_index(len(data), offsets, base_ind, burn_in)
signals = {'mid': mid, 'book': book, 'spr': spr, 'last': last}
signals.update(dict(zip(['bkema'+dn for dn in decay_names], emabook)))
signals.update(dict(zip(['vwap'+dn for dn in decay_names], vwap)))
signals.update(dict(zip(['trends'+dn for dn in decay_names], trends)))
signals.update(dict(zip(['trsign'+dn for dn in decay_names], trsigns)))
signals.update(dict(zip(['trsvolume'+dn for dn in decay_names], trsvolumes)))
signals.update(dict(zip(['signedv'+dn for dn in decay_names], sv)))
return signals, base_ind, offsets_ind
def get_correlation(signals, base_ind, offsets_ind):
futures = [signals['mid'][offind] - signals['mid'][base_ind] for offind in offsets_ind]
new_signals = {}
for key in signals.keys():
if key == 'mid' or key == 'spr':
continue
elif key == 'book' or 'bkema' in key or key == 'last' or 'vwap' in key:
new_signals[key] = (signals[key] - signals['mid'])[base_ind]
elif 'trends' in key:
new_signals[key] = (signals[key] - 1.0)[base_ind]
else :
new_signals[key] = signals[key][base_ind]
def get_cor_impl(sig, futures):
result = [np.mean(sig * fut) / np.std(sig) / np.std(fut) for fut in futures]
return result
correlation = {}
for key in new_signals.keys():
correlation[key] = get_cor_impl(new_signals[key], futures)
return correlation
#correlations = get_correlation(signals, base_ind, offsets_ind)
def print_correlations(correlations):
correlation_avg = [(key, correlations[key].mean(axis=0) * 100) for key in correlations.keys()]
correlation_max = [(key, correlations[key].mean(axis=0).mean() * 100) for key in correlations.keys()]
correlation_max = sorted(correlation_max, key = lambda x : x[1])
for signal, value in correlation_max:
print signal, '\t', value
print
trend = [item for item in correlation_max if 'trend' in item[0]]
book = [item for item in correlation_max if 'book' in item[0]]
last = [item for item in correlation_max if 'last' in item[0]]
emabook = [item for item in correlation_max if 'bkema' in item[0]]
vwap = [item for item in correlation_max if 'vwap' in item[0]]
trsign = [item for item in correlation_max if 'trsign' in item[0]]
trsvolume = [item for item in correlation_max if 'trsvolume' in item[0]]
signedv = [item for item in correlation_max if 'signedv' in item[0]]
print 'Trend signals'
print trend[0]
print dict(correlation_avg)[trend[0][0]]
print trend[-1]
print dict(correlation_avg)[trend[-1][0]]
print
print 'book signals'
print book[0]
print dict(correlation_avg)[book[0][0]]
#print trend[-1]
#print dict(correlation_avg)[trend[-1][0]]
print
print 'last signals'
print last[0]
print dict(correlation_avg)[last[0][0]]
#print trend[-1]
#print dict(correlation_avg)[trend[-1][0]]
print
print 'emabook signals'
print emabook[0]
print dict(correlation_avg)[emabook[0][0]]
print emabook[-1]
print dict(correlation_avg)[emabook[-1][0]]
print
print 'vwap signals'
print vwap[0]
print dict(correlation_avg)[vwap[0][0]]
print vwap[-1]
print dict(correlation_avg)[vwap[-1][0]]
print
print 'Trsign signals'
print trsign[0]
print dict(correlation_avg)[trsign[0][0]]
print trsign[-1]
print dict(correlation_avg)[trsign[-1][0]]
print
print 'Trsvolume signals'
print trsvolume[0]
print dict(correlation_avg)[trsvolume[0][0]]
print trsvolume[-1]
print dict(correlation_avg)[trsvolume[-1][0]]
print
print 'Signedv signals'
print signedv[0]
print dict(correlation_avg)[signedv[0][0]]
print signedv[-1]
print dict(correlation_avg)[signedv[-1][0]]
print
correlations = defaultdict(list)
for adate in dates:
print "processing %s"%adate
data = result[adate]
signals, base_ind, offsets_ind = get_signals(data)
tmp_cor = get_correlation(signals, base_ind, offsets_ind)
for key, value in tmp_cor.items():
correlations[key].append(value)
for key in correlations.keys():
correlations[key] = np.array(correlations[key])
print 'DONE!'
print_correlations(correlations)