Make forecast with GARCH models
Previously you have implemented a basic GARCH(1,1) model with the Python arch
package. In this exercise, you will practice making a basic volatility forecast.
You will again use the historical returns of S&P 500 time series. First define and fit a GARCH(1,1) model with all available observations, then call .forecast()
to make a prediction. By default it produces a 1-step ahead estimate. You can use horizon = n
to specify longer forward periods.
The arch
package has been preloaded for you.
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.
# Specify a GARCH(1,1) model
basic_gm = ____(sp_data['Return'], p = 1, q = 1,
mean = 'constant', vol = 'GARCH', dist = 'normal')
# Fit the model
gm_result = basic_gm.____()