Diagnostic summary statistics
It is important to know when you need to go back to the drawing board in model design. In this exercise you will use the residual test statistics in the results summary to decide whether a model is a good fit to a time series.
Here is a reminder of the tests in the model summary:
Test | Null hypothesis | P-value name |
---|---|---|
Ljung-Box | There are no correlations in the residual |
Prob(Q) |
Jarque-Bera | The residuals are normally distributed | Prob(JB) |
An unknown time series df
and the ARIMA
model class are available for you in your environment.
This exercise is part of the course
ARIMA Models in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create and fit model
model1 = ARIMA(df, order=____)
results1 = model1.fit()
# Print summary
print(____)