Guía para la importación de series

financieras de yahoo finanzas y la

aplicación del modelo VAR en
Python
 Modelo VARIMAX
Autor: Sergio Diaz Paredes

Contacto: Linkedin

Paquetes
In [ ]: pip install pmdarima

Collecting pmdarima
Downloading pmdarima-2.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_6
4.manylinux_2_28_x86_64.whl (2.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.1/2.1 MB 8.9 MB/s eta 0:00:00
Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.10/dist-pack
ages (from pmdarima) (1.3.2)
Requirement already satisfied: Cython!=0.29.18,!=0.29.31,>=0.29 in /usr/local/lib/
python3.10/dist-packages (from pmdarima) (3.0.8)
Requirement already satisfied: numpy>=1.21.2 in /usr/local/lib/python3.10/dist-pac
kages (from pmdarima) (1.25.2)
Requirement already satisfied: pandas>=0.19 in /usr/local/lib/python3.10/dist-pack
ages (from pmdarima) (1.5.3)
Requirement already satisfied: scikit-learn>=0.22 in /usr/local/lib/python3.10/dis
t-packages (from pmdarima) (1.2.2)
Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-pack
ages (from pmdarima) (1.11.4)
Requirement already satisfied: statsmodels>=0.13.2 in /usr/local/lib/python3.10/di
st-packages (from pmdarima) (0.14.1)
Requirement already satisfied: urllib3 in /usr/local/lib/python3.10/dist-packages
(from pmdarima) (2.0.7)
Requirement already satisfied: setuptools!=50.0.0,>=38.6.0 in /usr/local/lib/pytho
n3.10/dist-packages (from pmdarima) (67.7.2)
Requirement already satisfied: packaging>=17.1 in /usr/local/lib/python3.10/dist-p
ackages (from pmdarima) (23.2)
Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.1
0/dist-packages (from pandas>=0.19->pmdarima) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-pack
ages (from pandas>=0.19->pmdarima) (2023.4)
Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/d
ist-packages (from scikit-learn>=0.22->pmdarima) (3.3.0)
Requirement already satisfied: patsy>=0.5.4 in /usr/local/lib/python3.10/dist-pack
ages (from statsmodels>=0.13.2->pmdarima) (0.5.6)
Requirement already satisfied: six in /usr/local/lib/python3.10/dist-packages (fro
m patsy>=0.5.4->statsmodels>=0.13.2->pmdarima) (1.16.0)
Installing collected packages: pmdarima
Successfully installed pmdarima-2.0.4

In [ ]: import yfinance as yf
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from statsmodels.tsa.statespace.varmax import VARMAX

/usr/local/lib/python3.10/dist-packages/yfinance/base.py:48: FutureWarning: The de

fault dtype for empty Series will be 'object' instead of 'float64' in a future ver
sion. Specify a dtype explicitly to silence this warning.
_empty_series = pd.Series()
 Importando datos de yahoo finanzas
In [ ]: symbols = ["SONY", "DIS", "AAPL","AMZN"]

In [ ]: precio_cierre = pd.DataFrame()

In [ ]: for symbol in symbols:

stock_data = yf.download(symbol, start="2023-01-01")
precio_cierre[symbol] = stock_data['Close']

[*********************100%%**********************] 1 of 1 completed
[*********************100%%**********************] 1 of 1 completed
[*********************100%%**********************] 1 of 1 completed
[*********************100%%**********************] 1 of 1 completed

Extrapolando datos
In [ ]: precio_cierre = precio_cierre.resample('D').interpolate(method='akima')

In [ ]: precio_cierre.head(10)

Out[ ]: SONY DIS AAPL AMZN

Date

2023-01-03 76.650002 88.970001 125.070000 85.820000

2023-01-04 77.800003 91.980003 126.360001 85.139999

2023-01-05 77.500000 91.919998 125.019997 83.120003

2023-01-06 81.389999 93.919998 129.619995 86.080002

2023-01-07 82.203787 94.242645 129.861623 86.615078

2023-01-08 82.588781 94.339376 129.915749 86.652015

2023-01-09 82.620003 94.769997 130.149994 87.360001

2023-01-10 82.080002 95.559998 130.729996 89.870003

2023-01-11 84.290001 96.330002 133.490005 95.089996

2023-01-12 86.070000 99.809998 133.410004 95.269997

Analisis exploratorio
In [ ]: precio_cierre.plot(subplots=True)

array([<Axes: xlabel='Date'>, <Axes: xlabel='Date'>,

Out[ ]:
<Axes: xlabel='Date'>, <Axes: xlabel='Date'>], dtype=object)
In [ ]: precio_cierre.diff().plot(subplots=True)

array([<Axes: xlabel='Date'>, <Axes: xlabel='Date'>,

Out[ ]:
<Axes: xlabel='Date'>, <Axes: xlabel='Date'>], dtype=object)

In [ ]: precio_cierre.diff().corr()
Out[ ]: SONY DIS AAPL AMZN

SONY 1.000000 0.198643 0.410610 0.383595

DIS 0.198643 1.000000 0.291077 0.228025

AAPL 0.410610 0.291077 1.000000 0.396992

AMZN 0.383595 0.228025 0.396992 1.000000

In [ ]: sns.pairplot(precio_cierre.diff())

<seaborn.axisgrid.PairGrid at 0x7d4f40273670>
Out[ ]:

Partición de los datos

In [ ]: n_test = 14
df_train = precio_cierre.iloc[:-n_test]
df_test = precio_cierre.iloc[-n_test:]

Estimando el modelo
In [ ]: model = VARMAX(df_train, order=(7,1,7))
results = model.fit(disp=False)

/usr/local/lib/python3.10/dist-packages/statsmodels/tsa/statespace/varmax.py:161:
EstimationWarning: Estimation of VARMA(p,q) models is not generically robust, due
especially to identification issues.
warn('Estimation of VARMA(p,q) models is not generically robust,'
/usr/local/lib/python3.10/dist-packages/statsmodels/base/model.py:607: Convergence
Warning: Maximum Likelihood optimization failed to converge. Check mle_retvals
warnings.warn("Maximum Likelihood optimization failed to "

Forecast
In [ ]: forecast = results.get_forecast(steps=n_test)
forecast_df = pd.DataFrame(forecast.predicted_mean, columns=df_train.columns)

In [ ]: forecast_df.set_index(df_test.index,inplace=True)

In [ ]: forecast_df.head()

Out[ ]: SONY DIS AAPL AMZN

Date

2024-02-09 95.836284 110.871851 187.366744 169.401321

2024-02-10 96.452802 111.003489 188.420902 169.560694

2024-02-11 96.039634 110.085178 187.445026 168.739673

2024-02-12 96.054413 108.694129 187.087810 168.773534

2024-02-13 95.904601 108.966770 187.075437 169.054313

In [ ]: forecast_df.plot(subplots=True)

array([<Axes: xlabel='Date'>, <Axes: xlabel='Date'>,

Out[ ]:
<Axes: xlabel='Date'>, <Axes: xlabel='Date'>], dtype=object)
In [ ]: fig,ax = plt.subplots(2,2,figsize=(16,9))
#Train
i=0
for x in range(0,2):
for y in range(0,2):
#Train
ax[x,y].plot(df_train[symbols[i]],label='Train')
ax[x,y].legend()

ax[x,y].plot(df_test[symbols[i]],label='Test')
ax[x,y].legend()

ax[x,y].plot(forecast_df[symbols[i]],label='Forecast')
ax[x,y].legend()

ax[x,y].set_title(symbols[i])
i+=1
plt.show()