I need to estimate GARCH-M in state space form to find time varying risk aversion. The model is this:
Where r is the return of any asset.
I tried to estimate this model in Eviews, using Kalman filter, for the return of any asset, but the model always returns an error. There is an error in EViews that it can't estimate a state space model with nonlinear errors in state equation.
My questions are :
1 - There is a way to linearize this model ?
2 - In this case, I tried use Kalman Filter as in Chou and Engle, 1992. But there are other aproaches ?
3 - There is a R or Python library to estimate this model ?
I see a similar question here but the link, to a possible library, that the individual who answered the question provided does not work.
I apreciate any help.