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.
