Pick a winner based on AIC/BIC
In this exercise, you will practice using information criteria to choose a model with the best fit.
Information criteria intend to measure the trade-off between goodness of fit and model complexity. AIC and BIC are two commonly used information criteria for model selection. They both impose penalties on model with more parameters, or more complex models. The lower the AIC or BIC, the better the model.
A GJR-GARCH model and EGARCH model have been defined and fitted with the S&P 500 return data. Their results can be accessed in gjrgm_result
and egarch_result
respectively.
This exercise is part of the course
GARCH Models in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Print the AIC GJR-GARCH
print('AIC of GJR-GARCH model :', ____.____)
# Print the AIC of EGARCH
print('AIC of EGARCH model :', ____.____)
# Print the BIC GJR-GARCH
print('BIC of GJR-GARCH model :', ____.____)
# Print the BIC of EGARCH
print('BIC of EGARCH model :', ____.____)