金融时间序列#

In [1]: import numpy as np
import pandas as pd
from pylab import mpl, plt
plt.style.use('seaborn')
mpl.rcParams['font.family'] = 'serif'
%matplotlib inline

In [2]: filename = '.././source/tr_eikon_eod_data.csv'
In [3]: f = open(filename, 'r') ②
f.readlines()[:5] ②
Out[3]: ['Date,AAPL.0,MSFT.0,INTC.0,AMZN.0,GS.N,SPY,.SPX,.VIX,EUR=,XAU=,GDX, ,GLD\n',
'2010-01-01,,,,,,,1.4323,1096.35,,\n',
'2010-01-04,30.57282657,30.95,20.88,133.9,173.08,113.33,1132.99,20.04,
,1.4411,1120.0,47.71,109.8\n',
'2010-01-05,30.625683660000004,30.96,20.87,134.69,176.14,113.63,1136.52,
,19.35,1.4368,1118.65,48.17,109.7\n',
'2010-01-06,30.138541290000003,30.77,20.8,132.25,174.26,113.71,1137.14,
,19.16,1.4412,1138.5,49.34,111.51\n']

In [4]: data = pd.read_csv(filename, index_col=0, parse_dates=True)
In [5]: data.info() 6
<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 2216 entries, 2010-01-01 to 2018-06-29
Data columns (total 12 columns):
AAPL.02138 non-null float64
MSFT.02138 non-null float64
INTC.02138 non-null float64
AMZN.02138 non-null float64
GS.N 2138 non-null float64
SPY 2138 non-null float64
.SPX 2138 non-null float64
.VIX 2138 non-null float64
EUR= 2216 non-null float64
XAU= 2211 non-null float64
GDX 2138 non-null float64
GLD 2138 non-null float64
dtypes: float64(12)
memory usage: 225.1 KB

In [7]: data.tail()

In [8]: data.plot(figsize=(10, 12), subplots=True);

In [9]: instruments = ['Apple Stock', 'Microsoft Stock', 'Intel Stock', 'Amazon Stock', 'Goldman Sachs Stock', 'SPDR S&P 500 ETF Trust', 'S&P 500 Index', 'VIX Volatility Index', 'EUR/USD Exchange Rate', 'Gold Price', 'VanEck Vectors Gold Miners ETF', 'SPDR Gold Trust']

In [10]: for ric, name in zip(data.columns, instruments):
    print('{:8s} | {}'.format(ric, name))
AAPL.O | Apple Stock
MSFT.O | Microsoft Stock
INTC.O | Intel Stock
AMZN.O | Amazon Stock
GS.N | Goldman Sachs Stock
SPY | SPDR S&P 500 ETF Trust
.SPX | S&P 500 Index
.VIX | VIX Volatility Index
EUR= | EUR/USD Exchange Rate
XAU= | Gold Price
GDX | VanEck Vectors Gold Miners ETF
GLD | SPDR Gold Trust

In [11]: data.info() ①
<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 2216 entries, 2010-01-01 to 2018-06-29
Data columns (total 12 columns):
AAPL.O 2138 non-null float64
MSFT.O 2138 non-null float64
INTC.O 2138 non-null float64
AMZN.O 2138 non-null float64
GS.N 2138 non-null float64
SPY 2138 non-null float64
.SPX 2138 non-null float64
.VIX 2138 non-null float64
EUR= 2216 non-null float64
XAU= 2211 non-null float64
GDX 2138 non-null float64
GLD 2138 non-null float64
dtypes: float64(12)
memory usage: 225.1 KB

In [13]: data.mean( ) ①
Out[13]: AAPL.O 93.455973
MSFT.O 44.561115
INTC.O 29.364192
AMZN.O 480.461251
GS.N 170.216221
SPY 180.323029
.SPX 1802.713106
.VIX 17.027133
EUR= 1.248587
XAU= 1349.014130
GDX 33.566525
GLD 130.086590
dtype: float64

In [14]: data.aggregate([min, ②
np.mean, ③
np.std, ④
np.median, ⑤
max] ⑥
).round(2)

Out[14]:
AAPL.O MSFT.O INTC.O AMZN.O GS.N SPY .SPX .VIX EUR= \
min 27.4423.0117.66108.6187.70102.201022.589.141.04
mean 93.4644.5629.36480.46170.22180.321802.7117.031.25
std 40.5519.538.17372.3142.4848.19483.345.880.11
median 90.5539.6627.33322.06164.43186.321863.0815.581.27

In [15]: data.diff().head() ①
Out[15]:

In [16]: data.diff().mean()
Out[16]: AAPL.00.064737
MSFT.00.031246
INTC.00.013540
AMZN.00.706608
GS.N 0.028224
SPY 0.072103
.SPX 0.732659
.VIX -0.019583
EUR= -0.000119
XAU= 0.041887
GDX -0.015071
GLD -0.003455
dtype: float64

In [17]: data.pct_change().round(3).head()
Out[17]:

AAPL.O MSFT.O INTC.O AMZN.O GS.N SPY .SPX .VIX EUR= \
Date
2010-01-01 NaN NaN NaN NaN NaN NaN NaN NaN NaN 0.0062010-01-04 NaN NaN NaN NaN NaN NaN NaN NaN 0.003 -0.034 -0.0032010-01-050.0020.000 -0.0000.0060.0180.0030.003 -0.034 -0.0032010-01-06 -0.016 -0.006 -0.003 -0.018 -0.0110.0010.001 -0.0100.0032010-01-07 -0.002 -0.010 -0.010 -0.0170.0200.0040.004 -0.005 -0.007
XAU= GDX GLD
Date
2010-01-01 NaN NaN NaN
2010-01-040.022 NaN NaN
2010-01-05 -0.0010.010 -0.0012010-01-060.0180.0240.0162010-01-07 -0.006 -0.005 -0.006

In [18]: data.pct_change().mean().plot(kind='bar', figsize=(10, 6));

In [19]: rets = np.log(data / data.shift(1)) ①
In [20]: rets.head().round(3) ②
Out[20]:
AAPL.O MSFT.O INTC.O AMZN.O GS.N SPY .SPX .VIX EUR= \
Date
2010-01-01 NaN NaN NaN NaN NaN NaN NaN NaN NaN
2010-01-04 NaN NaN NaN NaN NaN NaN NaN NaN 0.0062010-01-050.0020.000 -0.0000.0060.0180.0030.003 -0.035 -0.0032010-01-06 -0.016 -0.006 -0.003 -0.018 -0.0110.0010.001 -0.0100.0032010-01-07 -0.002 -0.010 -0.010 -0.0170.0190.0040.004 -0.005 -0.007
XAU= GDX GLD
Date
2010-01-01 NaN NaN NaN
2010-01-040.021 NaN NaN
2010-01-05 -0.0010.010 -0.001

2010-01-060.0180.0240.0162010-01-07 -0.006 -0.005 -0.006

In [21]: rets.cumsum().apply(np.exp).plot(figsize=(10, 6));

In [23]: data.resample('1m', label='right').last().head()

In [24]: rets.cumsum().apply(np.exp).resample('1m', label='right').last().plot(figsize=(10, 6));

In [25]: sym = 'AAPL.O'

In [26]: data = pd.DataFrame(data[sym]).dropna()

In [27]: data.tail()

Out[27]: AAPL.O AAPL.0

Date 2018-06-25182.172018-06-26184.432018-06-27184.162018-06-28185.502018-06-29185.11

In [28]: window = 20

In [29]: data['min'] = data[sym].rolling(window=window).min()

In [30]: data['mean'] = data[sym].rolling(window=window).mean()

In [31]: data['std'] = data[sym].rolling(window=window).std()

In [32]: data['median'] = data[sym].rolling(window=window).median()

In [33]: data['max'] = data[sym].rolling(window=window).max()

In [34]: data['ewma'] = data[sym].ewm(halflife=0.5, min_periods=window).mean()

In [35]: data.dropna().head()

ewma
Date
2010-02-0127.8054322010-02-0227.9363372010-02-0328.3301342010-02-0427.6592992010-02-0527.856947

In [36]: ax = data[['min', 'mean', 'max']].iloc[-200:].plot( figsize=(10, 6), style=['g--', 'r--', 'g--'], lw=0.8) data[sym].iloc[-200:].plot(ax=ax, lw=2.0);

In [37]: data['SMA1'] = data[sym].rolling(window=42).mean()

In [38]: data['SMA2'] = data[sym].rolling(window=252).mean()

In [39]: data[[sym, 'SMA1', 'SMA2']].tail()
Out[39]: AAPL.0 SMA1 SMA2
Date
2018-06-25182.17185.606190168.2655562018-06-26184.43186.087381168.4187702018-06-27184.16186.607381168.5792062018-06-28185.50187.089286168.7366272018-06-29185.11187.470476168.901032

In [40]: data[[sym, 'SMA1', 'SMA2']].plot(figsize=(10, 6));

In [42]: data['positions'] = np.where(data['SMA1'] > data['SMA2'], 1, -1)

In [41]: data.dropna(inplace=True)

In [43]: ax = data[[sym, 'SMA1', 'SMA2', 'positions']].plot(figsize=(10, 6), secondary_y='positions')
ax.get_legend().set_bbox_to_anchor((0.25, 0.85));

In [44]: raw = pd.read_csv('../source/tr_eikon_eod_data.csv', index_col=0, parse_dates=True)

In [45]: data = raw[['.SPX', '.VIX']].dropna()

In [47]: data.plot(subplots=True, figsize=(10, 6));

In [48]: data.loc[:'2012-12-31'].plot(secondary_y='.VIX', figsize=(10, 6));

In [49]: rets = np.log(data / data.shift(1))
In [50]: rets.head()
Out[50]: .SPX .VIX
Date
2010-01-04 NaN NaN
2010-01-050.003111 -0.0350382010-01-060.000545 -0.0098682010-01-070.003993 -0.0052332010-01-080.002878 -0.050024

In [51]: rets.dropna(inplace=True)

In [52]: rets.plot(subplots=True, figsize=(10, 6));

In [53]: pd.plotting.scatter_matrix(rets, 1
    alpha=0.2, 2
    diagonal='hist', 3
    hist_kwargs={'bins': 35}, 4
    figsize=(10, 6));

In [54]: reg = np.polyfit(rets['.SPX'], rets['.VIX'], deg=1)
In [55]: ax = rets.plot(kind='scatter', x='.SPX', y='.VIX', figsize=(10, 6))
ax.plot(rets['.SPX'], np.polyval(reg, rets['.SPX']), 'r', lw=2);

In [56]: rets.corr() 1
Out[56]: .SPX .VIX
. SPX 1.000000 -0.804382
. VIX -0.8043821.000000

In [57]: ax = rets['.SPX'].rolling(window=252).corr(rets['.VIX']).plot(figsize=(10, 6)) ax.axhline(rets.corr().iloc[0, 1], c='r');

In [59]: %%time
    # data from FXCM Forex Capital Markets Ltd.
    tick = pd.read_csv('../source/fxcm_eur_usd_tick_data.csv',
    index_col=0, parse_dates=True)
    CPU times: user 1.07 s, sys: 149 ms, total: 1.22 s
    Wall time: 1.16 s

In [60]: tick.info()
    <class 'pandas.core.frame.DataFrame'>
    DatetimeIndex: 461357 entries, 2018-06-2900:00:00.082000 to 2018-06-2920:59:00.607000
    Data columns (total 2 columns):
    Bid    461357 non-null float64
    Ask    461357 non-null float64
    dtypes: float64(2)
    memory usage: 10.6 MB

In [61]: tick['Mid'] = tick.mean(axis=1) ①

In [62]: tick['Mid'].plot(figsize=(10, 6));

In [63]: tick_resam = tick.resample(rule='5min', label='right').last()

In [65]: tick_resam['Mid'].plot(figsize=(10, 6));