Diagnostics
You have arrived at the model diagnostic stage. So far you have found that the initial time series was stationary, but may have one outlying point. You identified promising model orders using the ACF and PACF and confirmed these insights by training a lot of models and using the AIC and BIC.
You found that the ARMA(1,2) model was the best fit to our data and now you want to check over the predictions it makes before you would move it into production.
The time series savings
has been loaded and the ARIMA
class has been imported into your environment.
This exercise is part of the course
ARIMA Models in Python
Exercise instructions
- Retrain the ARMA(1,2) model on the time series, setting the trend to constant.
- Create the 4 standard diagnostics plots.
- Print the model residual summary statistics.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create and fit model
model = ____
results = ____
# Create the 4 diagostics plots
____
plt.show()
# Print summary
____