optimizer.py
3.09 KB
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
import datetime
import pandas as pd
import numpy as np
import FinanceDataReader as fdr
from scipy.optimize import minimize
import json
class c_Models:
#Input 값으로, 자산 list, 사용자 포트폴리오 비중, 시작일, 마지막일
def __init__(self, assets, assets_w, start, end):
self.result = None
self.graph = None
data = pd.DataFrame()
# 전체 자산 data들을 가지고 온 후, 정리함
for asset in assets: #total_list:
tmp = fdr.DataReader(asset,start,end).Close
tmp.rename(columns={'Close': asset}, inplace=True)
data = pd.concat([data, tmp], axis=1)
if data.isnull().values.any() == True: #불러온 data에 오류가 있다면
return "No Data",''
else:
data = data.resample('M').mean() #일별 데이터를 월별 데이터로 만들어줌
data = data.pct_change() #월별 주가 데이터를 이용해 수익률 데이터로 변환
data.dropna(inplace=True) #결측치 제외(첫 row)
self.data = data
self.assets_w = assets_w
self.mu = data.mean() * 12
self.cov = data.cov() * 12
#GMV 최적화 : 제약 조건은 비중합=1, 공매도 불가능
def gmv_opt(self):
n_assets = len(self.data.columns)
w0 = np.ones(n_assets) / n_assets
fun = lambda w: np.dot(w.T, np.dot(self.cov, w))
constraints = ({'type':'eq', 'fun':lambda x: np.sum(x)-1})
bd = ((0,1),) * n_assets
#cov = data.cov() * 12
gmv = minimize(fun, w0, method = 'SLSQP', constraints=constraints, bounds=bd)
return gmv.x
#Max Sharp ratio : risk free rate은 0.8%로 지정했고,
def ms_opt(self):
n_assets = len(self.data.columns)
w0 = np.ones(n_assets) / n_assets
fun = lambda w: -(np.dot(w.T, self.mu) - 0.008) / np.sqrt(np.dot(w.T, np.dot(self.cov, w)))
bd = ((0,1),) * n_assets
constraints = ({'type': 'eq', 'fun': lambda x: np.sum(x) - 1})
maxsharp = minimize(fun, w0, method ='SLSQP', constraints=constraints, bounds=bd)
return maxsharp.x
def rp_opt(self):
def RC(cov, w):
pfo_std = np.sqrt(np.dot(w.T, np.dot(self.cov, w)))
mrc = 1/pfo_std * (np.dot(self.cov, w))
rc = mrc * w
rc = rc / rc.sum()
return rc
def RP_objective(x):
pfo_std = np.sqrt(np.dot(x.T, np.dot(self.cov, x)))
mrc = 1/pfo_std * (np.dot(self.cov, x))
rc = mrc * x
rc = rc / rc.sum()
a = np.reshape(rc, (len(rc),1))
differs = a - a.T
objective = np.sum(np.square(differs))
return objective
n_assets = len(self.data.columns)
w0 = np.ones(n_assets) / n_assets
constraints = [{'type':'eq', 'fun': lambda x: np.sum(x) -1}]
bd = ((0,1),) * n_assets
rp = minimize(RP_objective, w0, constraints=constraints, bounds = bd, method='SLSQP')
return rp.x #, RC(self.cov, rp.x)