I'm evaluating the impact of two variables on stock returns. For this I am using a Garch(1,1)-model in RStudio. This is the result I am getting. Why is the garch model not a valid choice? The external regressors are in the variace model, because they are not incorporated in the mean equation.

> myspec<-ugarchspec(variance.model=list(model="sGarch", garchOrder=c(1,1),external.regressors=mydata),
+                    mean.model=list(armaOrder=c(1,0),include.mean=TRUE))
ugarchspec-->error: the garch model does not appear to be a valid choice.
  • $\begingroup$ Apparently it is because the sgarch has to be written as sGARCH. $\endgroup$ – femma Oct 31 '17 at 13:51
  • 4
    $\begingroup$ I'm voting to close this question as off-topic because I do not see how it could be on topic. $\endgroup$ – Richard Hardy Nov 1 '17 at 20:28

From the docs:

List containing the variance model specification: model Valid models (currently implemented) are “sGARCH”, “fGARCH”, “eGARCH”, “gjrGARCH”, “apARCH” and “iGARCH” and “csGARCH”.

Other values than the ones listed are never valid.


Not the answer you're looking for? Browse other questions tagged or ask your own question.