[R-SIG-Finance] rugarch convergence problem

[Cabot_Bear]

Hi, Alexios

I am having convergence problem when fit a few time series using GARCH
family models. The system is continuously reporting errors i.e. solver
failer to converge or nonvalid 'nrow' value (too large or NA), depending on
the length of dataset. Some previous discussions suggested various solvers,
hence "nlminb", "gosolnp" and "nloptr" have been considered but non of them
can resolve this problem. One typical example can be found here including a
dataset and commands
https://www.dropbox.com/sh/lu8haexfz254eab/AAAstEhd9ywWuovKp8FLm5STa

Many many thanks,

Marco



--
View this message in context: http://r.789695.n4.nabble.com/rugarch-convergence-problem-tp4694878.html
Sent from the Rmetrics mailing list archive at Nabble.com.