drift/exploration.ipynb

In [2]:

from load_data import load_files
import pandas as pd
from pandas.plotting import lag_plot
import seaborn as sns
import matplotlib.pyplot as plt

data = load_files('data', True)
data = data[[column for column in data.columns if not column.endswith('volume')]]
data = data[[column for column in data.columns if column.startswith('BTC')]]
data.info()

<class 'pandas.core.frame.DataFrame'>
Index: 1500 entries, 2017-10-03 to 2021-11-10
Data columns (total 9 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   BTC_returns  1500 non-null   float64
 1   BTC_vol_10   1500 non-null   float64
 2   BTC_vol_20   1500 non-null   float64
 3   BTC_vol_30   1500 non-null   float64
 4   BTC_mom_10   1500 non-null   float64
 5   BTC_mom_20   1500 non-null   float64
 6   BTC_mom_30   1500 non-null   float64
 7   BTC_mom_60   1500 non-null   float64
 8   BTC_mom_90   1500 non-null   float64
dtypes: float64(9)
memory usage: 117.2+ KB

In [3]:

data.head(5)

Out [3]:

	BTC_returns	BTC_vol_10	BTC_vol_20	BTC_vol_30	BTC_mom_10	BTC_mom_20	BTC_mom_30	BTC_mom_60	BTC_mom_90
2017-10-03	-0.019799	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
2017-10-04	-0.022141	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
2017-10-05	0.024363	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
2017-10-06	0.011686	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
2017-10-07	0.014609	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0

In [3]:

lag_plot(data['SPY_returns'])

Out [3]:

<AxesSubplot:xlabel='y(t)', ylabel='y(t + 1)'>

In [4]:

lag_plot(data['BTC_returns'])

Out [4]:

<AxesSubplot:xlabel='y(t)', ylabel='y(t + 1)'>

In [10]:

pd.plotting.scatter_matrix(data, figsize=(12, 12));

In [6]:

fig, ax = plt.subplots(figsize=(20,20))
sns.heatmap(data.corr(), annot=True, ax=ax);

Out [6]:

<AxesSubplot:>

In [ ]: