Hi, It simply means that the model failed to converge to a global optimum. Usually this is caught before post-estimation calculations place so that the routine exits more gracefully, but in this case the problem was 'flagged' during the calculation of the robust standard errors which contained some NAs (likely some parameters where at their extreme boundary values).
Please send me a reproducible example of this offline (with data) so that I may add another layer of checks to avoid such errors. Regards, Alexios On 6 Feb 2014, at 10:20 AM, tvernay <[email protected]> wrote: > Hello, > > I have the following error which sometimes comes out from the ugarchfit > function (rugarch package): > > Error in diag(fit$robust.cvar) : nonvalid 'nrow' value (too large or NA) > > This error only appears for certain data sets but is not systematic even for > data sets of exactly the same size. > > This error disappears when I use less external regressors > (external.regressors input parameter in ugarchspec function). I would like > to understand better the origin of this error and how I can remedy to this > problem while keeping all my external regressors. > > Thanks a lot, cheers, > > Thibaut > > > > -- > View this message in context: > http://r.789695.n4.nabble.com/understanding-an-error-from-ugarchfit-tp4684836.html > Sent from the Rmetrics mailing list archive at Nabble.com. > > _______________________________________________ > [email protected] 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. > _______________________________________________ [email protected] 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.
