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.
Cet exercice fait partie du cours
GARCH Models in Python
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de 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.____()