[R-SIG-Finance] rcoplua.gauss sim problem

Joe W. Byers ecjbosu at aol.com
Thu Sep 6 12:29:31 CEST 2007


I may not be the best for answering your questions, but will try. 
garchFit may allow exogenous variable in the mean function thru the 
parameter xreg (see arima and arimaFit for more information).  The 
variance equation in garchFit, to my knowledge, is limited to garch/arch 
variance specification based on references in the help files.

I might recommend gamlss package for your question.  It allows 
specifications of exogenous variables in the variance equations and will 
allow you to fit many different distributions to your data.  It provides 
AIC stats for model specification tests.  I used gamlss in the example.

Also, be patient with rmetrics, they will respond.  I am still waiting 
on help with the rcopula question.

Good luck

Rich Ghazarian wrote:
 > Joe,
 > I am new to R and enjoy reading your posting on [R-SIG-Finance] .  I 
am looking at your code and my limited ability is not allowing me to 
help you with your problem, but I find it an excellent learning tool.  I 
have a question related to a GARCH estimation problem, looking at your 
code I am trying to figure out if it is possible to impose seasonality 
into the garchFit() or any function.  I would like monthly dummies to 
enter the garch process via alpha,.gamma, and  beta.   I know this is 
done in Splus, but can't seem to figure it out in  R.  Given your 
experience in R and Energy markets, I am hoping that you might have 
encountered such problem in the past.
 >  I have tried to contact the owner of rmetrics, but I have not heard 
anything.   Your help is much appreciated
 > Best,
 > Rich Ghazarian

More information about the R-SIG-Finance mailing list