-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathPortfolio.py
executable file
·197 lines (188 loc) · 7.3 KB
/
Portfolio.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
from copy import deepcopy
import matplotlib.pyplot as plt
import pandas as pd
from Stock import Stock
from TimeSeries import TimeSeries
from utility import draw_zero_axis, plot_cauchy, plot_gaussian
class Portfolio:
def __init__(self, stocks: list, metrics=[Stock.RETURN, Stock.LOG_RETURN], max_values:int=10000):
for stock in stocks:
stock.time_series.df = stock.time_series.df.head(max_values)
self.stocks = stocks
self.symbols = [stock.symbol for stock in stocks]
self.reference_stock: Stock = stocks[0]
time_vector = self.reference_stock.time_series.df[Stock.DATE]
self.metrics = {}
for metric in metrics:
self.metrics[metric] = {Stock.DATE: time_vector}
for stock in stocks:
stock: Stock
self.metrics[metric][stock.symbol] = stock.time_series.df[metric]
self.metrics[metric] = TimeSeries(
pd.DataFrame(self.metrics[metric]), Stock.DATE
)
def plot_time_series(self):
fig = plt.figure()
ax1 = fig.add_subplot(3, 1, 1)
draw_zero_axis(ax1)
for stock in self.stocks:
stock: Stock
stock.plot_time_series(
ax1, Stock.CLOSE, label=f"Price ({Stock.UNITS})", legend=True
)
ax2 = fig.add_subplot(3, 1, 2)
draw_zero_axis(ax2)
for stock in self.stocks:
stock.plot_time_series(ax2, Stock.RETURN)
ax3 = fig.add_subplot(3, 1, 3)
draw_zero_axis(ax3)
for stock in self.stocks:
stock.plot_time_series(ax3, Stock.LOG_RETURN)
fig.suptitle("Time Series")
return fig
def plot_distributions(self):
fig = plt.figure()
legend = True
for figure_idx, metric in enumerate([Stock.RETURN, Stock.LOG_RETURN], 1):
ax = fig.add_subplot(2, 1, figure_idx)
if figure_idx != 1:
legend = False
means = []
volatilities = []
for stock in self.stocks:
stock: Stock
stock.plot_distribution(ax, metric)
means.append(stock.mean(metric))
volatilities.append(stock.volatility(metric))
avg_mean = sum(means) / len(means)
avg_volatility = sum(volatilities) / len(volatilities)
plot_gaussian(ax, avg_mean, avg_volatility)
plot_cauchy(ax, avg_mean, 0.25 * avg_volatility)
ax.set_xlim([-0.25, 0.25])
ax.set_ylim([0, 100])
if legend:
ax.legend()
fig.suptitle("Distributions\n(Non-Gaussian)")
return fig
def plot_statistics_over_time(self, window_size: int, metric=Stock.LOG_RETURN):
"""Plot mean and volatility over a rolling window (window_size in days)."""
fig = plt.figure()
ax1 = fig.add_subplot(1, 1, 1)
for stock in self.stocks:
stock: Stock
stock.plot_time_series(
ax1, metric, alpha=0.3, kwargs_overrides={"linewidth": 0.5}
)
common_kwargs = {
"drawstyle": "steps",
"xlabel": stock.DATE,
"ylabel": f"{metric} (log-{stock.UNITS})",
"alpha": 0.7,
"legend": True,
"color": stock.color,
"grid": True,
}
mean_ts = stock.mean_over_time(metric, window_size)
mean_kwargs = deepcopy(common_kwargs)
mean_kwargs.update(
{
"label": f"{stock.company} Mean",
"linewidth": 2,
}
)
mean_ts.plot(metric, ax1, mean_kwargs)
volatility_ts = stock.volatility_over_time(metric, window_size)
volatility_kwargs = deepcopy(common_kwargs)
volatility_kwargs.update(
{
"label": f"{stock.company} Volatility",
"linewidth": 1,
}
)
volatility_ts.plot(metric, ax1, volatility_kwargs)
ax1.legend()
fig.suptitle(
"Non-Stationary Mean & Volatility of Log-Returns\n"
f"({window_size}-day rolling statistics)"
)
return fig
def compare_stocks(self, metric=Stock.LOG_RETURN, window_size=365):
if len(self.stocks) < 2:
raise Exception("Portfolio must contain at least 2 stocks to compare.")
metric_ts: TimeSeries = self.metrics[metric]
reference_metric = metric_ts.df[self.reference_stock.symbol]
reference_mean_ts = self.reference_stock.mean_over_time(metric, window_size)
reference_volatility_ts = self.reference_stock.volatility_over_time(
metric, window_size
)
fig = plt.figure()
ax1 = fig.add_subplot(3, 1, 1)
draw_zero_axis(ax1)
difference_kwargs = {
"ylabel": f"{u'Δ'}{metric} (USD)",
"ylim": [-0.1, 0.1],
"alpha": 0.3,
}
ax2 = fig.add_subplot(3, 1, 2)
draw_zero_axis(ax2)
mean_difference_kwargs = {
"ylabel": f"Mean\n{u'Δ'}{metric} (USD)",
"ylim": [-0.01, 0.01],
"alpha": 0.7,
}
ax3 = fig.add_subplot(3, 1, 3)
draw_zero_axis(ax3)
volatility_difference_kwargs = {
"ylabel": f"Volatility\n{u'Δ'}{metric} (USD)",
"ylim": [-0.01, 0.01],
"alpha": 0.7,
}
for stock in self.stocks[1:]:
stock: Stock
label = f"{stock.symbol}-{self.reference_stock.symbol}"
mean_ts = stock.mean_over_time(metric, window_size)
volatility_ts = stock.volatility_over_time(metric, window_size)
difference_ts = TimeSeries(
pd.DataFrame(
{
Stock.DATE: metric_ts.df[Stock.DATE],
label: metric_ts.df[stock.symbol] - reference_metric,
}
),
Stock.DATE,
)
mean_difference_ts = TimeSeries(
pd.DataFrame(
{
Stock.DATE: mean_ts.df[Stock.DATE],
label: mean_ts.df[metric] - reference_mean_ts.df[metric],
}
),
Stock.DATE,
)
volatility_difference_ts = TimeSeries(
pd.DataFrame(
{
Stock.DATE: volatility_ts.df[Stock.DATE],
label: volatility_ts.df[metric]
- reference_volatility_ts.df[metric],
}
),
Stock.DATE,
)
common_kwargs = {
"label": label,
"color": stock.color,
"xlabel": Stock.DATE,
"grid": True,
}
difference_kwargs.update(common_kwargs)
mean_difference_kwargs.update(common_kwargs)
volatility_difference_kwargs.update(common_kwargs)
difference_ts.plot(label, ax1, difference_kwargs)
mean_difference_ts.plot(label, ax2, mean_difference_kwargs)
volatility_difference_ts.plot(label, ax3, volatility_difference_kwargs)
ax1.legend()
ax2.legend()
ax3.legend()
return fig