-
Notifications
You must be signed in to change notification settings - Fork 8
/
MA_Strategy.py
203 lines (169 loc) · 9.94 KB
/
MA_Strategy.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
# -*- coding: utf-8 -*-
"""
Created on Fri Apr 10 08:20:15 2020
@author: yuba316
"""
import sys
sys.path.append(r'D:\work\back_test_system')
import DataBase as DB
import BackTest as BT
import copy as c
import math as m
import numpy as np
import pandas as pd
import datetime
#%% 获得基于标的物所生成的交易信号
ma,devU,devD,observer = 40,0.03,0.03,5
flag = 1/observer*m.ceil(observer/2)
df = DB.getUnderlying(end_date='20200410')[['trade_date','close','pre_close']]
df['ma'] = df['pre_close'].rolling(window=ma).mean()
df['signal'] = df.apply(lambda x: 1 if x['pre_close']>=x['ma']*(1+devU) else (-1 if x['pre_close']<=x['ma']*(1-devD) else 0),axis=1)
df['observer'] = df['signal'].rolling(window=observer).mean()
df['observer'] = df['observer'].apply(lambda x: 1 if x>=flag else (-1 if x<=-1*flag else 0))
df.dropna(inplace=True)
df.reset_index(drop=True,inplace=True)
#%% 获取每月的期权到期时间
start_date = df['trade_date'].iloc[0]
end_date = df['trade_date'].iloc[-2]
option_basic = DB.getOpBasic()
maturity_date = list(option_basic[(option_basic['maturity_date']>start_date)&(option_basic['maturity_date']<end_date)]['maturity_date'].unique())
maturity_date.append(list(option_basic[option_basic['maturity_date']>maturity_date[-1]]['maturity_date'].unique())[0])
#%% 生成开平仓信号
option_maturity_date,order,direction = [],[],[]
n = len(df)
option_maturity_date.append(maturity_date[0])
order.append(2)
direction.append(df['signal'].iloc[0])
i=1
while(i<n-1):
if (i+3<n-1) and df['trade_date'].iloc[i+3] in maturity_date: # 还有4天到期时开仓
option_maturity_date.append(maturity_date[maturity_date.index(df['trade_date'].iloc[i+3])+1])
order.append(2)
direction.append(df['signal'].iloc[i])
elif (i+4<n-1) and df['trade_date'].iloc[i+4] in maturity_date: # 还有5天到期时平仓
option_maturity_date.append(option_maturity_date[-1])
order.append(-2)
direction.append(direction[-1])
else:
option_maturity_date.append(option_maturity_date[-1])
direction.append(direction[-1])
last_cover_date = maturity_date[maturity_date.index(option_maturity_date[-1])-1]
if (i>4) and (i+6<n-1) and (df['trade_date'].iloc[i-4] > last_cover_date) and (df['trade_date'].iloc[i+5] < option_maturity_date[-1])\
and (direction[-1]!=df['observer'].iloc[i]): # 开仓已过去5天，且距离平仓还有2天时间，开启监视哨
order.append(-2)
option_maturity_date.append(option_maturity_date[-1])
order.append(2)
direction.append(df['signal'].iloc[i])
i = i+2
continue
order.append(0)
i = i+1
option_maturity_date.append(option_maturity_date[-1])
order.append(-2)
direction.append(direction[-1])
df['maturity_date'] = option_maturity_date
df['order'] = order
df['direction'] = direction
#%% 合并期权合约行情
n = len(df)
df['call'],df['put'] = np.nan,np.nan
for i in range(n):
if df['order'].iloc[i]==2:
option = c.deepcopy(option_basic[(option_basic['maturity_date']==df['maturity_date'].iloc[i])&\
(option_basic['list_date']<=df['trade_date'].iloc[i])])
option['distance'] = abs(option['exercise_price']-df['pre_close'].iloc[i])
call = option[option['call_put']=='C']
call_distance = list(call['distance'])
call_index = min(call_distance.index(min(call_distance))+1,len(call)-1)
df['call'].iloc[i] = call['ts_code'].iloc[call_index]
put = option[option['call_put']=='P']
put_distance = list(put['distance'])
put_index = max(put_distance.index(min(put_distance))-1,0)
df['put'].iloc[i] = put['ts_code'].iloc[put_index]
df.fillna(method='ffill',inplace=True)
code = list(df['call'].unique())+list(df['put'].unique())
OpDB = pd.read_csv(r'D:\work\back_test_system\DataBase\Option\OP510050.csv')
OpDB = OpDB[OpDB['ts_code'].apply(lambda x: x in code)]
OpDB['trade_date'] = OpDB['trade_date'].apply(str)
#%% 拆分成signalDf、depositDf和vixDf
# signalDf[DataFrame]: [trade_date, Call_close, Put_close, Call_volume, Put_volume, signal, direction, position, pct]
signalDf = c.deepcopy(df[['trade_date','close','pre_close','order','direction','call','put','maturity_date']])
signalDf.rename(columns={'close':'underlying_close','pre_close':'underlying_pre_close','order':'signal'},inplace=True)
signalDf['direction'] = signalDf['direction'].apply(lambda x: 0 if x==1 else(1 if x==-1 else 2))
signalDf.rename(columns={'call':'ts_code'},inplace=True)
signalDf = pd.merge(signalDf,OpDB[['ts_code','trade_date','close','pre_close','pre_settle','exercise_price']],how='left',on=['ts_code','trade_date'])
signalDf.rename(columns={'close':'Call_close','pre_close':'Call_pre_close','pre_settle':'Call_pre_settle','exercise_price':'Call_exercise'},inplace=True)
signalDf.rename(columns={'ts_code':'call'},inplace=True)
signalDf.rename(columns={'put':'ts_code'},inplace=True)
signalDf = pd.merge(signalDf,OpDB[['ts_code','trade_date','close','pre_close','pre_settle','exercise_price']],how='left',on=['ts_code','trade_date'])
signalDf.rename(columns={'close':'Put_close','pre_close':'Put_pre_close','pre_settle':'Put_pre_settle','exercise_price':'Put_exercise'},inplace=True)
signalDf.rename(columns={'ts_code':'put'},inplace=True)
# depositDf[DataFrame]: [trade_date, underlying_pre_close, Call_pre_close, Put_pre_close, Call_pre_settle, Put_pre_settle, Call_exercise, Put_exercise]
depositDf = c.deepcopy(signalDf['trade_date, underlying_pre_close, Call_pre_close, Put_pre_close, Call_pre_settle, Put_pre_settle, Call_exercise, Put_exercise'.split(', ')])
vixDf = c.deepcopy(signalDf[['trade_date','underlying_close','Call_close','Put_close','Call_exercise','Put_exercise','maturity_date']])
signalDf = signalDf[['trade_date','signal','direction','Call_close','Put_close']]
#%% 计算隐含波动率，准备加减仓
rfDf = DB.getRf(vixDf['trade_date'].iloc[0],vixDf['trade_date'].iloc[-1])
vixDf = pd.merge(vixDf,rfDf,how='left',on='trade_date')
vixDf['T'] = vixDf.apply(lambda x: ((datetime.datetime.strptime(x['maturity_date'],'%Y%m%d')-\
datetime.datetime.strptime(x['trade_date'],'%Y%m%d')).days+1)/365,axis=1)
vixDf['CorP'] = 1
vixDf['Call_vix'] = vixDf.apply(lambda x: BT.CalVIX(x['underlying_close'],x['Call_exercise'],x['Call_close'],x['rf'],x['T'],x['CorP']),axis=1)
vixDf['CorP'] = 0
vixDf['Put_vix'] = vixDf.apply(lambda x: BT.CalVIX(x['underlying_close'],x['Put_exercise'],x['Put_close'],x['rf'],x['T'],x['CorP']),axis=1)
vixDf['Call_rng'] = vixDf['Call_vix']-vixDf['Call_vix'].shift(1)
vixDf['Put_rng'] = vixDf['Put_vix']-vixDf['Put_vix'].shift(1)
vixDf['Call_rng'] = vixDf['Call_rng'].shift(1)
vixDf['Put_rng'] = vixDf['Put_rng'].shift(1)
vixDf.fillna(0,inplace=True)
signalDf = pd.merge(signalDf,vixDf[['trade_date','Call_rng','Put_rng']],how='left',on='trade_date')
#%% 加减仓
signalDf['position'] = 0
n = len(signalDf)
flagOp,flagCl = True,True
for i in range(n):
if signalDf['signal'].iloc[i]==0:
if flagOp and ((signalDf['direction'].iloc[i]==0 and signalDf['Put_rng'].iloc[i]>=0.05) or \
(signalDf['direction'].iloc[i]==1 and signalDf['Call_rng'].iloc[i]>=0.05) or \
(signalDf['direction'].iloc[i]==2 and signalDf['Call_rng'].iloc[i]>=0.05 and signalDf['Put_rng'].iloc[i]>=0.05)):
signalDf['position'].iloc[i] = 1
flagOp = False
elif flagCl and ((signalDf['direction'].iloc[i]==0 and signalDf['Put_rng'].iloc[i]<=-0.05) or \
(signalDf['direction'].iloc[i]==1 and signalDf['Call_rng'].iloc[i]<=-0.05) or \
(signalDf['direction'].iloc[i]==2 and (signalDf['Call_rng'].iloc[i]<=-0.05 or signalDf['Put_rng'].iloc[i]<=-0.05))):
signalDf['position'].iloc[i] = -1
flagCl = False
elif signalDf['signal'].iloc[i]==2:
flagOp,flagCl = True,True
signalDf['pct'] = 0.1
signalDf['Call_volume'] = -1
signalDf['Put_volume'] = -1
signalDf.drop(['Call_rng','Put_rng'],axis=1,inplace=True)
#%% 回测
signalDf['trade_date'] = signalDf['trade_date'].apply(lambda x: datetime.datetime.strptime(x,'%Y%m%d'))
depositDf['trade_date'] = depositDf['trade_date'].apply(lambda x: datetime.datetime.strptime(x,'%Y%m%d'))
depositDf = BT.getOpDeposit(depositDf,10000)
recordDf = BT.OptionBT(signalDf,depositDf)
benchmarkDf = c.deepcopy(df[['trade_date','close']])
benchmarkDf['trade_date'] = benchmarkDf['trade_date'].apply(lambda x: datetime.datetime.strptime(x,'%Y%m%d'))
stat = BT.Visualize(recordDf,benchmarkDf)
#%% 用 BackTest_2 测试
signalDf_2 = c.deepcopy(signalDf)
signalDf_2['signal'] = signalDf_2['signal'].apply(lambda x: 1 if x==2 else (-1 if x==-2 else 0))
signalDf_2['price'] = signalDf_2.apply(lambda x: [x['Call_close'],x['Put_close']] if x['direction']==2 else \
([x['Call_close']] if x['direction']==1 else [x['Put_close']]),axis=1)
signalDf_2['position'] = signalDf_2.apply(lambda x: [1,1] if ((x['direction']==2) and (x['position']==1)) else \
([-1,-1] if ((x['direction']==2) and (x['position']==-1)) else \
([1] if x['position']==1 else ([-1] if x['position']==-1 else \
([0,0] if x['direction']==2 else [0])))),axis=1)
signalDf_2['pct'] = signalDf_2['direction'].apply(lambda x: [0.1,0.1] if x==2 else [0.1])
signalDf_2 = pd.merge(signalDf_2,depositDf,how='left',on='trade_date')
signalDf_2['deposit'] = signalDf_2.apply(lambda x: [x['Call_dep'],x['Put_dep']] if x['direction']==2 else \
([x['Call_dep']] if x['direction']==1 else [x['Put_dep']]),axis=1)
signalDf_2['direction'] = signalDf_2['direction'].apply(lambda x: [-1,-1] if x==2 else [-1])
signalDf_2['volume'] = signalDf_2['direction']
signalDf_2.drop(['Call_close','Put_close','Call_volume','Put_volume','Call_dep','Put_dep'],axis=1,inplace=True)
#%%
import BackTest_2 as BT_2
recordDf_2 = BT_2.OptionBT(signalDf_2)
stat_2 = BT_2.Visualize(recordDf,benchmarkDf)