In my write-up for task 1, I was going to mention that the Regime-Switching Lognormal Model is a more conservative model because it displays a better ability to depict a fat tail distribution; however, when I measure kurtosis, I get 2.16 for the LN model versus 1.32 for the RSLN model. Wouldn't this indicate that the LN model has a fatter tail?
Task 1: Kurtosis
0 commentaires:
Enregistrer un commentaire