Plot diagnostics
It is important to know when you need to go back to the drawing board in model design. In this exercise you will use 4 common plots to decide whether a model is a good fit to some data.
Here is a reminder of what you would like to see in each of the plots for a model that fits well:
| Test | Good fit |
|---|---|
| Standardized residual | There are no obvious patterns in the residuals |
| Histogram plus kde estimate | The KDE curve should be very similar to the normal distribution |
| Normal Q-Q | Most of the data points should lie on the straight line |
| Correlogram | 95% of correlations for lag greater than zero should not be significant |
An unknown time series df and the ARIMA model class are available for you in your environment.-
Deze oefening maakt deel uit van de cursus
ARIMA Models in Python
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
# Create and fit model
model = ARIMA(df, order=(____))
results = model.fit()
# Create the 4 diagnostics plots
____
plt.show()