forked from jjasonn0717/pred_finance
-
Notifications
You must be signed in to change notification settings - Fork 0
/
save_stock.py
236 lines (213 loc) · 9.09 KB
/
save_stock.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
226
227
228
229
230
231
232
233
234
235
236
import sys
import time
from yahoo_finance import Share
import numpy as np
import datetime
import argparse
if __name__ == "__main__":
## argument parse ##
parser = argparse.ArgumentParser(description="Calc analysis and Saving data from Yahoo Finance!")
parser.add_argument("name", help="company name of applied index")
parser.add_argument("index", help="Stock index of the company")
parser.add_argument("--end",'-e', help="ending date, yyyy-mm-dd/today default to today", default='today')
parser.add_argument("--start", '-s', help="starting date, yyyy-mm-dd, default to 1990-1-1", default='1990-1-1')
parser.add_argument("--ratio", '-r', help="ratio of training data, default to 1.0, which stand for no splittng and only one data file will be generated",
type=float,
default=1.0)
args = parser.parse_args()
name = args.name
Stock = Share(args.index)
if args.end == "today":
end = str(datetime.date.today())
else:
end = args.end
start = args.start
if args.ratio < 0. or args.ratio > 1.:
raise "Invalid ratio argumant! Should be in range 0.~1."
else:
TRAIN_RATIO = args.ratio
history_data = Stock.get_historical(start, end)
history_data.reverse()
## delete data which volume equals zero ##
history_data = [data for data in history_data if not float(data['Volume']) == 0]
data_size = len(history_data)
daily_d_close = np.array([], dtype='float')
daily_RSI9 = np.array([], dtype='float')
daily_RSI15 = np.array([], dtype='float')
daily_VA_D = np.array([], dtype='float')
if TRAIN_RATIO == 1.0:
train_file = open('./stock_data/'+name+'.csv', 'w')
else:
train_file = open('./stock_data/'+name+'_train.csv', 'w')
test_file = open('./stock_data/'+name+'_test.csv', 'w')
print "Days in total:", data_size
print "slice data into..."
print "Train:", int(data_size*TRAIN_RATIO), " Test:", data_size - int(data_size*TRAIN_RATIO)
train_file.write(Stock.get_info()['symbol'] + ',' + 'High,Low,Open,Close,d_Close,RSI9,RSI15,MA5,MA20,MA60,d_CO,d_HL,Adj_Close,Volume,VA/D,d_VA/D,%R8,%R21,DIF,DEM,d_MA5/20\n')
if not TRAIN_RATIO == 1.0:
test_file.write(Stock.get_info()['symbol'] + ',' + 'High,Low,Open,Close,d_Close,RSI9,RSI15,MA5,MA20,MA60,d_CO,d_HL,Adj_Close,Volume,VA/D,d_VA/D,%R8,%R21,DIF,DEM,d_MA5/20\n')
VA_D_tm1 = 0.
EMA12_tm1 = 0.
EMA26_tm1 = 0.
DEM_tm1 = 0.
for i in xrange(data_size):
data = history_data[i]
volume = float(data['Volume'])
d_CO = float(data['Close']) - float(data['Open'])
d_HL = float(data['High']) - float(data['Low'])
d_CL = float(data['Close']) - float(data['Low'])
d_HC = float(data['High']) - float(data['Close'])
## calc d_close ##
if i == 0:
d_close = 0.
else:
d_close = float(data['Close']) - float(history_data[i-1]['Close'])
daily_d_close = np.append(daily_d_close, [d_close])
## calc RSI9 ##
if i < 9:
sig_abs_d_close = np.sum(np.absolute(daily_d_close))
sig_pos_d_close = np.sum(np.multiply(daily_d_close, (daily_d_close > 0)))
else:
sig_abs_d_close = np.sum(np.absolute(daily_d_close[i-9+1:]))
sig_pos_d_close = np.sum(np.multiply(daily_d_close[i-9+1:], (daily_d_close[i-9+1:] > 0)))
assert sig_abs_d_close >= 0 and sig_pos_d_close >= 0
if sig_abs_d_close == 0:
daily_RSI9 = np.append(daily_RSI9, [0.])
else:
daily_RSI9 = np.append(daily_RSI9, [sig_pos_d_close / sig_abs_d_close])
## calc RSI15 ##
if i < 15:
sig_abs_d_close = np.sum(np.absolute(daily_d_close))
sig_pos_d_close = np.sum(np.multiply(daily_d_close, (daily_d_close > 0)))
else:
sig_abs_d_close = np.sum(np.absolute(daily_d_close[i-15+1:]))
sig_pos_d_close = np.sum(np.multiply(daily_d_close[i-15+1:], (daily_d_close[i-15+1:] > 0)))
assert sig_abs_d_close >= 0 and sig_pos_d_close >= 0
if sig_abs_d_close == 0:
daily_RSI15 = np.append(daily_RSI15, [0.])
else:
daily_RSI15 = np.append(daily_RSI15, [sig_pos_d_close / sig_abs_d_close])
## calc MA5 ##
MA5 = 0.
count = 0.
for t in xrange(5):
if (i - t) >= 0:
MA5 += float(history_data[i - t]['Close'])
count += 1.
MA5 /= count
## calc MA20 ##
MA20 = 0.
count = 0.
for t in xrange(20):
if (i - t) >= 0:
MA20 += float(history_data[i - t]['Close'])
count += 1.
MA20 /= count
## calc MA60 ##
MA60 = 0.
count = 0.
for t in xrange(60):
if (i - t) >= 0:
MA60 += float(history_data[i - t]['Close'])
count += 1.
MA60 /= count
## calc VA/D ##
if i == 0:
VA_D = volume
else:
if not d_HL == 0:
VA_D = daily_VA_D[-1] + ((d_CL - d_HC) / d_HL) * volume
else:
VA_D = daily_VA_D[-1]
daily_VA_D = np.append(daily_VA_D, [VA_D])
if i == 0:
d_VA_D = 0.
VA_D_tm1 = VA_D
else:
d_VA_D = (VA_D-VA_D_tm1)/1000000
VA_D_tm1 = VA_D
## calc piR8 ##
if i < 8:
min_dclose = np.min(daily_d_close[:])
max_dclose = np.max(daily_d_close[:])
else:
min_dclose = np.min(daily_d_close[i-8+1:])
max_dclose = np.max(daily_d_close[i-8+1:])
piR8 = (d_close - max_dclose)/(max_dclose-min_dclose) if not (max_dclose - min_dclose) == 0. else 0.
## calc piR21 ##
if i < 21:
min_dclose = np.min(daily_d_close[:])
max_dclose = np.max(daily_d_close[:])
else:
min_dclose = np.min(daily_d_close[i-21+1:])
max_dclose = np.max(daily_d_close[i-21+1:])
piR21 = (d_close - max_dclose)/(max_dclose-min_dclose) if not (max_dclose - min_dclose) == 0. else 0.
## calc MACD ##
DI = (float(data['High']) + float(data['Low']) + 2*float(data['Close'])) / 4.
if i == 0:
EMA12 = DI
EMA26 = DI
else:
EMA12 = (11./13.)*EMA12_tm1 + (2./13.)*DI
EMA26 = (25./27.)*EMA26_tm1 + (2./27.)*DI
EMA12_tm1 = EMA12
EMA26_tm1 = EMA26
DIF = EMA12 - EMA26
if i == 0:
DEM = DIF
else:
DEM = (8./10.)*DEM_tm1 + (2./10.)*DIF
DEM_tm1 = DEM
if volume == 0:
raise "volume == 0", data['Date']
if i < int(data_size*TRAIN_RATIO):
train_file.write(data['Date'] + ',' +
data['High'] + ',' +
data['Low'] + ',' +
data['Open'] + ',' +
data['Close'] + ',' +
str(d_close) + ',' +
str(daily_RSI9[i]) + ',' +
str(daily_RSI15[i]) + ',' +
#str(daily_RSI9[i]-daily_RSI15[i]) + ',' +
str(MA5) + ',' +
str(MA20) + ',' +
str(MA60) + ',' +
str(d_CO) + ',' +
str(d_HL) + ',' +
data['Adj_Close'] + ',' +
str(volume/1000000) + ',' +
str(VA_D/1000000) + ',' +
str(d_VA_D) + ',' +
str(piR8) + ',' +
str(piR21) + ',' +
str(DIF) + ',' +
str(DEM) + ',' +
str(MA5-MA20) + '\n')
else:
test_file.write(data['Date'] + ',' +
data['High'] + ',' +
data['Low'] + ',' +
data['Open'] + ',' +
data['Close'] + ',' +
str(d_close) + ',' +
str(daily_RSI9[i]) + ',' +
str(daily_RSI15[i]) + ',' +
#str(daily_RSI9[i]-daily_RSI15[i]) + ',' +
str(MA5) + ',' +
str(MA20) + ',' +
str(MA60) + ',' +
str(d_CO) + ',' +
str(d_HL) + ',' +
data['Adj_Close'] + ',' +
str(volume/1000000) + ',' +
str(VA_D/1000000) + ',' +
str(d_VA_D) + ',' +
str(piR8) + ',' +
str(piR21) + ',' +
str(DIF) + ',' +
str(DEM) + ',' +
str(MA5-MA20) + '\n')
train_file.close()
if not TRAIN_RATIO == 1.0:
test_file.close()