HOW TO ESTIMATE A SPLINE-GARCH MODEL?
Posted: Wed Feb 01, 2012 9:49 am
Hi all,
Does anybody know how to implement the spline-garch model introduced by Engle in 2005 with eviews?
The model is define as:
y = mu + res res~N(0,h_t)
h_t=g_t * f_t
where
g_t= (1-beta_1-beta_2) + beta_1*(res_{t-1}^2)/f_{t-1} + beta_2*g_{t-1}
f_t=gamma_0*exp [gamma_1*t + gamma_2*t^2 + Sum (i=1,k) w_i*max(0,t-t_i) ]
f_t is a deterministic function which represent here a quadratic spline specified as the sum of k truncated quadratic basis functions.
A full definition of the model is available here: http://economics.ucr.edu/seminars/fall0 ... 2-9-05.pdf (page 4)
I want to implement this model by using the log-likehood object of eviews but i don't know how to do that.
Does anyone have an idea on how to estimate the parameters for spline garch please?
Thanks in advance,
Regards,
Does anybody know how to implement the spline-garch model introduced by Engle in 2005 with eviews?
The model is define as:
y = mu + res res~N(0,h_t)
h_t=g_t * f_t
where
g_t= (1-beta_1-beta_2) + beta_1*(res_{t-1}^2)/f_{t-1} + beta_2*g_{t-1}
f_t=gamma_0*exp [gamma_1*t + gamma_2*t^2 + Sum (i=1,k) w_i*max(0,t-t_i) ]
f_t is a deterministic function which represent here a quadratic spline specified as the sum of k truncated quadratic basis functions.
A full definition of the model is available here: http://economics.ucr.edu/seminars/fall0 ... 2-9-05.pdf (page 4)
I want to implement this model by using the log-likehood object of eviews but i don't know how to do that.
Does anyone have an idea on how to estimate the parameters for spline garch please?
Thanks in advance,
Regards,