Hello everybody,

for my thesis I've been attempting to estimate a model for governement
credit spread changes with relatively many external variables
(macroeconomic announcements).
(that should be both in mean equation and in variance equation). So far
I've used OxMetrics module GARCH ( www.*garch*.*org*/  ) using its
implementation of MaxSA optimizer (Simulated Annealing,
http://personal.vu.nl/c.s.bos/packages/maxsa/doc/maxsa.html)

However the algorithm only reaches weak convergence with somewhat unstable
results (There is an error of maximum function evaluations reached). I've
been thinking I might try rugarch. According to your opinion, can "rugarch"
optimizers handle many  regressors in the equation? (say 15-20 external
regressors in the mean and 8 in the variance eq, but those are not as
crucial)

Any ideas, comments or advice appreciated,

Vojtìch Pi¹tora

__________________________________________________
Some additional information:
models attempted in OxMetrics: univariate GARCH models (simple GARCH /
EGARCH), with variance targetting
y= something like this:
https://www.dropbox.com/s/kunphdo7l3e4eds/Boxplot_spreads_cz.pdf
(spread changes, many outliers)

x= something like this + controls (interest rates, exchange rates, stock
market movements, volatility indices)
https://www.dropbox.com/s/80r9yj1dx1vc170/surpises_de.pdf
(macroeconomic surprises measured in standard deviations - they are only
nonzero about once a month or even less)

some papers that use the same framework:
http://www.tandfonline.com/doi/abs/10.1080/00036846.2011.587775#.UwdBNYWndjA
http://www.sciencedirect.com/science/article/pii/S1566014106000707

        [[alternative HTML version deleted]]

_______________________________________________
[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.

Reply via email to