/
Final_project_tree_strategy.py
136 lines (95 loc) · 4.24 KB
/
Final_project_tree_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
'''回測實證'''
import numpy as np #載入套件
import pandas as pd
import matplotlib.pyplot as plt
from datetime import datetime, date, time, timezone
import time
start = time.time()
data_liq = pd.read_excel('data.xlsx') #讀入電子股資料
list1 = data_liq['證券代碼'].unique()
'''篩流動性，在運用策略從中選股'''
data_liq1= {} #求出2015-2020電子類股公司的平均市值
for i in list1:
a = data_liq['市值(百萬元)'][data_liq['證券代碼']== i].mean()
data_liq1[i] = a
data_liq2 = pd.DataFrame.from_dict(data_liq1, orient='index',columns=['mv'])
score1 = data_liq2.sort_values(by = ['mv'],ascending=False)
score1 = score1.iloc[150:] #只取前150家市值最大公司，故先列出剩下的再去除
list_select_drop = score1.index
data_liq_final = data_liq.set_index('證券代碼' , inplace=False)
data_liq_final = data_liq_final.drop(list_select_drop,axis=0) #去掉流動性差的股票
'''只取出2020年資料'''
data_liq_final["年月日"] = pd.to_datetime(data_liq_final["年月日"])
data_liq_final = data_liq_final[data_liq_final["年月日"].dt.year==2020] #只取出2020年資料
df_benchmark = pd.read_excel('benchmark.xlsx') #讀入大盤資料
df_benchmark["年月"] = pd.to_datetime(df_benchmark["年月"])
df_benchmark = df_benchmark[df_benchmark["年月"].dt.year==2020]
'''策略一: ROE>=7.095 & P/E>=9.095'''
ROE = data_liq_final['ROE－綜合損益']
ROE = ROE.dropna(axis=0, how='any') #去除空值
NN = []
for n in range(150):
NN.append(n*3)
print(NN)
ROE_F = pd.DataFrame() #算出2020年各公司平均ROE
for i in NN:
ROE1 = ROE[i:i+3].mean() #算平均一季的報酬率
dict_ROE = {'證券代碼':ROE.index[i],'ROE':[ROE1]} #變成字典再轉成df
ROE2 = pd.DataFrame(dict_ROE)
ROE_F = ROE_F.append(ROE2)
ROE_F = ROE_F[ROE_F['ROE']>7.095] #篩選出ROE大於7.095的個股
PE = data_liq_final['本益比-TSE']
NN2 = []
for n2 in range(150):
NN2.append(n2*12)
print(NN2)
PE_F = pd.DataFrame() #算出2020年各公司平均PE
for j in NN2:
PE1 = PE[j:j+12].mean() #算平均一季的報酬率
dict_PE = {'證券代碼':PE.index[j],'P/E':[PE1]} #變成字典再轉成df
PE2 = pd.DataFrame(dict_PE)
PE_F = PE_F.append(PE2)
PE_F = PE_F[PE_F['P/E']>9.095] #篩選出P/E大於9.095的個股
#將上述條件篩選出的股票取交集，取出42家公司
ROE_list = set(ROE_F['證券代碼']) #資料類型轉為set
PE_list = set(PE_F['證券代碼'])
asset = (ROE_list & PE_list)
res = [asset1 for asset1 in asset] #set轉成list
data_liq_final_s1 = data_liq_final.loc[res,['年月日','報酬率％_月']] #只留篩選出的股票
len(data_liq_final_s1.index.unique()) #檢查是否只剩下42檔股票
#製作樞紐分析表
strategy1 = pd.pivot_table(data_liq_final_s1,values='報酬率％_月',index=['年月日'],columns=data_liq_final_s1.index)
'''策略二: P/E>18.702'''
PE_F2 = PE_F[PE_F['P/E']>18.702] #篩選出P/E>=9.322 & P/E<=18.702的個股，共95檔股票
PE2_list = PE_F2['證券代碼']
data_liq_final_s2 = data_liq_final.loc[PE2_list,['年月日','報酬率％_月']]
len(data_liq_final_s2.index.unique())
#製作樞紐分析表
strategy2 = pd.pivot_table(data_liq_final_s2,values='報酬率％_月',index=['年月日'],columns=data_liq_final_s2.index)
'''回測2020年選股策略績效，與大盤做比較'''
bench = df_benchmark['報酬率％_月'].values.tolist()
s1_return = [] #比重一樣的加權平均各策略報酬率，再進行回測比較
s2_return = []
for v in range(12):
s1 = strategy1.iloc[v,:].mean()
s2 = strategy2.iloc[v,:].mean()
s1_return.append(s1)
s2_return.append(s2)
s1_returnc = [0,0,0,0,0,0,0,0,0,0,0,0]
s2_returnc = [0,0,0,0,0,0,0,0,0,0,0,0]
benchc = [0,0,0,0,0,0,0,0,0,0,0,0]
for w in range(12):
s1_returnc[w] = s1_return[w-1] + s1_return[w]
s2_returnc[w] = s2_return[w-1] + s2_return[w]
benchc[w] = bench[w-1] + bench[w]
fig, ax = plt.subplots() #回測績效圖
ax.plot(strategy1.index,s1_returnc,label='strategy1',linestyle=':')
ax.plot(strategy1.index,s2_returnc,label='strategy2',linestyle='--')
ax.plot(strategy1.index,benchc,label='benchmark')
fig.set_figwidth(15)
fig.tight_layout()
plt.xticks(size='small',rotation=45,fontsize=8)
plt.legend()
plt.show
end = time.time()
print(end-start)