CommencerCommencer gratuitement

AR or MA

In this exercise you will use the ACF and PACF to decide whether some data is best suited to an MA model or an AR model. Remember that selecting the right model order is of great importance to our predictions.

Remember that for different types of models we expect the following behavior in the ACF and PACF:

AR(p)MA(q)ARMA(p,q)
ACFTails offCuts off after lag qTails off
PACFCuts off after lag pTails offTails off

A time series with unknown properties, df is available for you in your environment.

Cet exercice fait partie du cours

ARIMA Models in Python

Afficher le cours

Exercice interactif pratique

Essayez cet exercice en complétant cet exemple de code.

# Import
from statsmodels.graphics.tsaplots import ____, ____

# Create figure
fig, (ax1, ax2) = plt.subplots(2,1, figsize=(12,8))
 
# Plot the ACF of df
____(____, lags=____, zero=False, ax=ax1)

# Plot the PACF of df
____(____, lags=____, zero=____, ax=ax2)

plt.show()
Modifier et exécuter le code