In this recipe, we used the same data as in the previous one, in order to compare the results of the ARCH and GARCH models. For more information on downloading data, please refer to Step 1 to 4 in the Explaining stock returns' volatility with ARCH models recipe.

Due to the convenience of the arch library, it was very easy to adjust the code used previously to fit the ARCH model. To estimate the GARCH model, we had to specify the type of volatility model we wanted to use, and set q=1.

For comparison's sake, we left the mean process as a zero-mean process.