[R-SIG-Finance] rugarch by processing with plyr

[alexios ghalanos]

Why would they? You have fixed them in the spec (data$XVAR) and are only
rolling the YVAR (which is the same for both subjects) by subject.

You probably want:

ddply(data, 'SUBJECT', function(x)
coef(ugarchfit(ugarchspec(mean.model=list(armaOrder=c(0,0),
external.regressors=data.matrix(x$XVAR)),variance.model=list(model='eGARCH')),
data=x$YVAR, solver='hybrid')))


Regards,

Alexios

On 23/09/2013 20:43, Geoffrey Smith wrote:
> I would like to estimate an EGARCH model by subject using the rugarch and
> plyr packages.  I am concerned that the external regressors named in the
> ugarchspec function do not roll forward by subject.  For example, in the
> code below, both subjects have the exact same dependent variables, but
> different independent variables.  Yet, the parameter estimates for both
> subjects are the exact same.  I think this is due to the independent
> variables not changing by subject.  Am I correct?  If so, how can I work
> around this?  Thank you very much.
> 
>     library(plyr)
>     library(rugarch)
> 
>     set.seed(1000)
> 
>     Y <- rnorm(200)
> 
>     data <- data.frame(SUBJECT=c(rep('Alice', 200), rep('Bart', 200)),
> YVAR=rep(Y, 02), XVAR=rnorm(400))
>     data
> 
>     spec <- ugarchspec(mean.model=list(armaOrder=c(00,00),
> external.regressors=data.matrix(data$XVAR), include.mean=TRUE),
> variance.model=list(model='eGARCH', garchOrder=c(01,01)),
> distribution.model='norm')
> 
>     ddply(data, .(SUBJECT), function(x) coef(ugarchfit(spec=spec,
> data=x$YVAR, solver='hybrid')))
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.
> 
>