RUGARCH bootstrap fitting error - presigma length

[stoyan.stoyanov]

Hi all,

While trying to bootstrap a TGARCH model of orders (1,1), using the rugarch package, I get the following error:

.fgarchsim2(fit = fit, n.sim = n.sim, n.start = n.start, m.sim = m.sim,  :
ugarchsim-->error: presigma must be of length 2

I only get the error when my ARMA model for the mean is of order 2 or higher (in p, q or both). Curious whether this is a bug that anyone has documented before. If not, I would be happy to provide my code and data sample. It could also be a mis-specification in my model and/or an error in the code of course.

Best,
Stoyan

[alexios]

Hi Stoyan,

You need to provide at a minimum the code you used. When you say
bootstrap which function are you referring to (ugarchsim or ugarchboot)?

In any case I cannot replicate the error under the following set of
assumptions:

library(rugarch)
data(sp500ret)
spec = ugarchspec(mean.model=list(armaOrder=c(3,2)),
variance.model=list(model="fGARCH",submodel="TGARCH"))
fit=ugarchfit(spec, sp500ret)

# These all work:
sim = ugarchsim(fit, n.sim = 1, m.sim=200, startMethod="unconditional")
sim = ugarchsim(fit, n.sim = 1, m.sim=200, startMethod="sample")
bt=ugarchboot(fit, method = "Partial", n.bootfit = 200)

(the use of .fgarchsim2 means that (n.sim+n.start) < 100 && m.sim >
100...this makes use of C++ code whereas the opposite case is based on C
code).

Regards,
Alexios

On 25/06/2012 22:58, stoyan.stoyanov wrote:

> Hi all,
>
> While trying to bootstrap a TGARCH model of orders (1,1), using the rugarch
> package, I get the following error:
>
> .fgarchsim2(fit = fit, n.sim = n.sim, n.start = n.start, m.sim = m.sim,  :
> ugarchsim-->error: presigma must be of length 2
>
> I only get the error when my ARMA model for the mean is of order 2 or higher
> (in p, q or both). Curious whether this is a bug that anyone has documented
> before. If not, I would be happy to provide my code and data sample. It
> could also be a mis-specification in my model and/or an error in the code of
> course.
>
> Best,
> Stoyan
>
> --
> View this message in context: http://r.789695.n4.nabble.com/RUGARCH-bootstrap-fitting-error-presigma-length-tp4634470.html
> Sent from the Rmetrics mailing list archive at Nabble.com.
>
> _______________________________________________
> [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.
>

_______________________________________________
[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.

[stoyan.stoyanov]

Hi Alexios,

I am using 1372 daily adjusted returns on IBM (csv attached).

Code:
spec=ugarchspec(variance.model = list(model = "fGARCH", garchOrder = c(3,3),
                        submodel = "TGARCH", external.regressors = NULL, variance.targeting = TRUE),
                 mean.model = list (armaOrder = c(best.p,best.q), include.mean = mean, archm = FALSE,
                        archpow = 1, arfima = FALSE, external.regressors = NULL, archex = FALSE),
                 distribution.model = "std", start.pars = list(), fixed.pars = list())

fit=ugarchfit(spec, data, out.sample = 0, solver = "solnp", solver.control = list(),
               fit.control = list(stationarity = 1, fixed.se = 0, scale = 0))

boot.pred=ugarchboot(fit, data = NULL, method = "full", n.ahead = 5,
                n.bootfit = 5, n.bootpred = 500, out.sample = 0, rseed = NA, solver = "solnp",
                solver.control = list(), fit.control = list(),
                external.forecasts = list(mregfor = NULL, vregfor = NULL), parallel = FALSE,
                parallel.control = list(pkg = c("multicore", "snowfall"), cores = 2))

Error:
fitting stage...done!

prediction stage...
Error in .fgarchsim2(fit = fit, n.sim = n.sim, n.start = n.start, m.sim = m.sim,  :
 
ugarchsim-->error: presigma must be of length 2

Again, I get the error with anything higher than an ARMA(1,1). Hope this information is sufficient. I get the same error with a different data set too.

Data:
data.csv

Thanks,
Stoyan

[alexios]

Thanks,

That's another bug caught and fixed (revision 424).

Regards,
Alexios

On 26/06/2012 13:53, stoyan.stoyanov wrote:

> Hi Alexios,
>
> I am using 1372 daily adjusted returns on IBM (csv attached).
>
> *Code:*
> spec=ugarchspec(variance.model = list(model = "fGARCH", garchOrder = c(3,3),
>                          submodel = "TGARCH", external.regressors = NULL,
> variance.targeting = TRUE),
>                   mean.model = list (armaOrder = c(best.p,best.q),
> include.mean = mean, archm = FALSE,
>                          archpow = 1, arfima = FALSE, external.regressors =
> NULL, archex = FALSE),
>                   distribution.model = "std", start.pars = list(), fixed.pars
> = list())
>
> fit=ugarchfit(spec, data, out.sample = 0, solver = "solnp", solver.control =
> list(),
>                 fit.control = list(stationarity = 1, fixed.se = 0, scale =
> 0))
>
> boot.pred=ugarchboot(fit, data = NULL, method = "full", n.ahead = 5,
>                  n.bootfit = 5, n.bootpred = 500, out.sample = 0, rseed = NA,
> solver = "solnp",
>                  solver.control = list(), fit.control = list(),
>                  external.forecasts = list(mregfor = NULL, vregfor = NULL),
> parallel = FALSE,
>                  parallel.control = list(pkg = c("multicore", "snowfall"),
> cores = 2))
>
> *Error:*
> fitting stage...done!
>
> prediction stage...
> Error in .fgarchsim2(fit = fit, n.sim = n.sim, n.start = n.start, m.sim =
> m.sim,  :
>
> ugarchsim-->error: presigma must be of length 2
>
> Again, I get the error with anything higher than an ARMA(1,1). Hope this
> information is sufficient. I get the same error with a different data set
> too.
>
> *Data:*
> http://r.789695.n4.nabble.com/file/n4634505/data.csv data.csv
>
> Thanks,
> Stoyan
>
> --
> View this message in context: http://r.789695.n4.nabble.com/RUGARCH-bootstrap-fitting-error-presigma-length-tp4634470p4634505.html
> Sent from the Rmetrics mailing list archive at Nabble.com.
>
> _______________________________________________
> [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.
>

_______________________________________________
[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.