[R-SIG-Finance] The ruGarch Package‏ & Insignificant estimates

[alexios ghalanos]

You have 3 options:

1. Fix the parameter which is not significant to zero and re-estimate
e.g. ma1
"setfixed(modelspec)<-list(ma1=0)"
The documentation on ugarchspec explains the naming of the parameters.

2. Change the ugarchspec to exclude the parameter from estimation.

3. Automatically have the program zero the parameters outside a certain
p-value threshold and re-estimate the model using the 'reduce' method:
e.g. given an estimated model of class uGARCHfit (e.g. 'mod')
newmod = reduce(mod, pvalue=0.1)

At present, the danger of using the 'reduce' method is that it will zero
the GARCH intercept if it is not significant. I will update this to
instead set it to its variance targeting value in the next update.

-Alexios

On 03/12/2013 02:58, Benny André Byremo wrote:
> Hello
> I hope somebody have time to answer me on a question related to the ruGarch package. I have fitted a model , but one of the estimates is insignificant and I want to estimate the same model once more without including this estimate. What should I type to make this happened? I have consulted the package pdf and googled it but without luck.
> The code I used:
> modelspec <- ugarchspec (variance.model = list(model = "gjrGARCH", garchOrder = c(1,1), submodel=0 ), mean.model = list(armaOrder = c(0, 1), include.mean = T), distribution.model = "norm" , fixed.pars=list())  modelspec at model$parsmodelfit <- ugarchfit(spec=modelspec, data=rn)show(model fit)  
> Greetings  		 	   		  
> 	[[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.
> 
>