Page 1 of 1
STAR- STGARCH Lundberg and Terasvirta (1999)
Posted: Tue Oct 18, 2011 10:41 am
by aymenbelgacem
Hi everyone
Is there any replication code for the STAR STGARCH model, as explained in Lundberg and Terasvirta, 1999 "Modelling economic high frequency time series". Stockholm school of economics WP
Re: STAR- STGARCH Lundberg and Terasvirta (1999)
Posted: Tue Oct 18, 2011 11:31 am
by TomDoan
aymenbelgacem wrote:Hi everyone
Is there any replication code for the STAR STGARCH model, as explained in Lundberg and Terasvirta, 1999 "Modelling economic high frequency time series". Stockholm school of economics WP
The TSAYP185.RPF example does a STAR-ARCH model. STAR-GARCH isn't that much different; you just have to use the standard recursive formulas for generating the variances as is done with other GARCH model estimated using MAXIMIZE.
Re: STAR- STGARCH Lundberg and Terasvirta (1999)
Posted: Tue Oct 18, 2011 2:48 pm
by aymenbelgacem
It's STAR STGARCH and not STAR GARCH (smooth transition in mean and variance equation)
Re: STAR- STGARCH Lundberg and Terasvirta (1999)
Posted: Wed Oct 19, 2011 9:18 pm
by TomDoan
You asked whether anyone had RATS code to do the analysis in a twelve year old unpublished working paper. The answer is almost certainly no. There are many reasons a paper might not have been published, but a good bet is that this wasn't considered particularly novel. It just combines a STAR model for the regression with a STAR model for the GARCH.