[R-SIG-Finance] NAs for GARCH

Brian G. Peterson brian at braverock.com
Tue Mar 13 02:59:36 CET 2007


On Monday 12 March 2007 20:42, Rich Ghazarian wrote:
> Here is what my data looks like, after I try to remove the NAs, but I
> still get an error message for GARCH, any suggestion as to how I can
> index this.  The data is irregular daily returns, and I would like to
> fit a GARCH type model, but I am not sure how to apply zooreg or ts

As I said in an earlier post, you have a few options. zoo provides a 
number of functions which may help you decide what to do with missing 
days.  na.locf, na.omit,  na.approx, or thw aggregate function may be 
used.  You'll need to decide which method is most appropriate to your 
data and best represents your missing observations.  If enough 
observations are missing, for instance, you may need to use aggregate to 
fit a weekly regular series.  Only examination of your data will tell you 
what the right answer is, the list can be of limited assistance.

Regards,

   - Brian

-- 
773-459-4973 mobile
http://braverock.com/brian/resume-quant.pdf



More information about the R-SIG-Finance mailing list