rugarch and gosolnp

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

rugarch and gosolnp

Geoffrey Smith-3
Hello, I am trying to estimate the EGARCH on many time series using the
code below.  Setting solver = "solnp" produces a lot of warnings().
Setting solver = "gosolnp" works perfectly but I get different parameter
estimates if I change rseed = in the solver.control list.  Is there any way
to make the "gosolnp" estimates less sensitive to the rseed = value?  In
other words, can I change some setting so that the "gosolnp" estimates do
not vary much based on different rseed = values?  Lastly, if I only care
about the parameter estimates (not the standard errors), can I just safely
ignore the warnings() and just use solver = "solnp"?  Thank you.

ugarchspec(mean.model = list(armaOrder = c(1, 1), include.mean = TRUE),
variance.model = list(model = "eGARCH", garchOrder = c(1, 1))),
solver = "gosolnp",
solver.control = list(rseed = 1234))

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] 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.