Exercise

# AIC and BIC vs ACF and PACF

In this exercise you will apply an AIC-BIC order search for the earthquakes time series. In the last lesson you decided that this dataset looked like an AR(1) process. You will do a grid search over parameters to see if you get the same results. The ACF and PACF plots for this dataset are shown below.

The `SARIMAX`

model class and the time series DataFrame `earthquake`

are available in your environment.

Instructions

**100 XP**

- Loop over orders of
`p`

and`q`

between 0 and 2. - Inside the loop
`try`

to fit an ARMA(p,q) to`earthquake`

on each loop. - Print
`p`

and`q`

alongside AIC and BIC in each loop. - If the model fitting procedure fails print
`p`

,`q`

,`None`

,`None`

.