Fixed rolling window forecast
Rolling-window forecasts are very popular for financial time series modeling. In this exercise, you will practice how to implement GARCH model forecasts with a fixed rolling window.
First define the window size inside .fit()
, and perform the forecast with a for-loop. Note since the window size remains fixed, both the start and end points increment after each iteration.
The S&P 500 return series has been preloaded as sp_data
, and a GARCH(1,1) model has been predefined in basic_gm
. The start and end points of the initial sample window has been pre-defined in start_loc
and end_loc
respectively.
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.
for i in range(30):
# Specify fixed rolling window size for model fitting
gm_result = basic_gm.fit(first_obs = i + ____,
last_obs = i + ____, update_freq = 5)