research/backtesting.py at master · faisalburhanudin/research · GitHub

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
# %%
from datetime import datetime

import talib
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt


# %%


def get_stock(symbol: str, start: str = None, end: str = None) -> pd.DataFrame:
    """Get Stock data frame

    Args:
        symbol: stock symbol
        start: start date trading, ex 2019-02-28
        end: end date trading, ex 2019-02-28

    Returns:
        DataFrame Stock
    """

    # download stock from github
    # raw data from repository
    df = pd.read_csv(f'https://raw.githubusercontent.com/faisalburhanudin/idx/master/stocks/{symbol}.csv')

    # convert string Date to DateTime
    df['DateTime'] = pd.DatetimeIndex(df['Date'])
    df['Month'] = pd.DatetimeIndex(df['Date']).month
    df['Year'] = pd.DatetimeIndex(df['Date']).year
    df['Day'] = pd.DatetimeIndex(df['Date']).day

    # limit dataframe start until start date
    if start:
        df = df[df['DateTime'] >= datetime.fromisoformat(start)]

    # limit dataframe until end date
    if end:
        df = df[df['DateTime'] <= datetime.fromisoformat(end)]

    return df


stock = get_stock('BBRI', '2000-01-01')


# %%
def buy_and_hold(stock_df: pd.DataFrame):
    money = 0

    record = []
    df_month = stock_df.groupby(['Year', 'Month'], as_index=False).first()
    for value in df_month.itertuples():
        # deposit money
        money += 1_000_000

        # price per lot
        per_lot = value.Close * 100

        # total share buy
        buy_share = money // per_lot * 100

        # total cost buy
        buy_cost = buy_share * value.Close

        # deduct from money
        money -= buy_cost

        record.append((value.DateTime, money, value.Close, buy_share))

    record = pd.DataFrame(record, columns=['DateTime', 'Money', 'Close', 'Share'])

    record['SumShare'] = record['Share'].expanding().sum()
    record['Value'] = record['SumShare'] * record['Close'] + record['Money']
    return record


record_df = buy_and_hold(stock)


# %%
def golden(stock_df: pd.DataFrame):
    stock_df['MA50'] = talib.SMA(stock_df['Close'], timeperiod=50)
    stock_df['MA200'] = talib.SMA(stock_df['Close'], timeperiod=200)

    money = 0
    record = []
    df_month = stock_df.groupby(['Year', 'Month'], as_index=False).first()
    for value in df_month.itertuples():
        # deposit money
        money += 1_000_000
        buy_share = 0

        if not np.isnan(value.MA50) and not np.isnan(value.MA200):

            # price per lot
            per_lot = value.Close * 100

            # buy stock
            if value.MA50 >= value.MA200:
                # total share buy
                buy_share = money // per_lot * 100

                # total cost buy
                buy_cost = buy_share * value.Close

                # deduct from money
                money -= buy_cost

            # sell stock
            else:
                # calculated total owned share
                own_share = get_owned_share([i[3] for i in record])
                if own_share:
                    buy_share = -own_share
                    money += own_share * value.Close

        record.append((value.DateTime, money, value.Close, buy_share, value.MA50, value.MA200))

    record = pd.DataFrame(record, columns=['DateTime', 'Money', 'Close', 'Share', 'MA50', 'MA200'])

    record['SumShare'] = record['Share'].expanding().sum()
    record['Value'] = record['SumShare'] * record['Close'] + record['Money']
    return record


def get_owned_share(buy_share):
    owned = 0
    for i in reversed(buy_share):
        if i < 0:
            break
        else:
            owned += i

    return owned


golden_df = golden(stock)

# %%
final = pd.DataFrame({
    'DateTime': record_df['DateTime'],
    'Close': record_df['Close'],
    'hold': record_df['SumShare'],
    'gold': golden_df['SumShare'],
    'reality': record_df['Value'],
    'idea': golden_df['Value'],
})

final.plot.line(x='DateTime', y=['reality', 'idea'])
ax = plt.gca()
ax.ticklabel_format(style='plain', axis='y')
plt.show()