Rugarch - ugarchroll (eGarch 1, 1) - conditional sigma results "inf" on skewed student t
Bernhard, Sorry, but I can't replicate the error. I attach the result of my run. Here is my sessionInfo() R version 3.4.3 (2017-11-30) Platform: x86_64-apple-darwin15.6.0 (64-bit) Running under: macOS High Sierra 10.13.4 Matrix products: default BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib LAPACK: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRlapack.dylib locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8 attached base packages: [1] parallel? stats???? graphics? grDevices utils???? datasets methods?? base other attached packages: [1] rugarch_1.4-0 loaded via a namespace (and not attached): ?[1] Rcpp_0.12.16??????????????? magrittr_1.5 knitr_1.16????????????????? misc3d_0.8-4 ?[5] xtable_1.8-2??????????????? lattice_0.20-35 R6_2.2.2??????????????????? FNN_1.1 ?[9] Rsolnp_1.16???????????????? GeneralizedHyperbolic_0.8-1 SkewHyperbolic_0.3-2??????? tools_3.4.3 [13] xts_0.10-2????????????????? spd_2.0-1 grid_3.4.3????????????????? KernSmooth_2.23-15 [17] htmltools_0.3.6???????????? yaml_2.1.18 digest_0.6.15?????????????? rgl_0.98.1 [21] numDeriv_2016.8-1?????????? Matrix_1.2-12 shiny_1.0.5???????????????? nloptr_1.0.4 [25] DistributionUtils_0.5-1???? ks_1.10.7 htmlwidgets_1.0???????????? codetools_0.2-15 [29] mime_0.5??????????????????? compiler_3.4.3 multicool_0.1-10??????????? expm_0.999-2 [33] jsonlite_1.5??????????????? truncnorm_1.0-7 mvtnorm_1.0-6?????????????? httpuv_1.3.6.2 [37] zoo_1.8-1 Regards, Alexios
On 4/17/18 2:16 PM, Bernhard Lange wrote:
Dear all
I?m trying to do a rolling forecast for a time series of stock returns
aiming to achieve One-step-ahead VaR99 forecasts with the rugarch package.
However, I'm facing problems while using the ugarchroll function.
When running this code:
library(rugarch)
specEGARCHNoMeanSstd <- ugarchspec(variance.model =
list(model="eGARCH", garchOrder = c(1,1)),
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?mean.model = list(armaOrder=c(0,0)),
distribution.model = "sstd")
EGarchNoMSstd <- ugarchroll(specEGARCHNoMeanSstd, data=returnsXTS,
n.start = 500, refit.every = 100,
? ? ? ? ? ? ? refit.window = "moving", window.size = 500,
solver="hybrid", calculate.VaR = T, VaR.alpha = c(0.01))
I receive the following Error as value in the resulting data object:
Error in try(.C("c_qsstd", p = as.double(p), mu = as.double(mu), sigma
= as.double(sigma), ?: \n ?NA/NaN/Inf in foreign function call (arg 3)\n.
I figured out that the function produces "Inf"s on the conditional
sigma. However I'm not able to find a reason for that.
Please note that the same model with n.start = 1000 and window.size =
1000 produces data without errors.
I would appreciate every hint in the right direction.
I attached a data set and script for reproduction.
Best regards,
Bernhard Lange
PS: I?m new to R mailing lists. So please let me know, if I need to
provide more information for future postings.
_______________________________________________ R-SIG-Finance at r-project.org 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.
-------------- next part -------------- An HTML attachment was scrubbed... URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20180417/9d379b71/attachment.html> -------------- next part -------------- A non-text attachment was scrubbed... Name: testdata2.rdata Type: application/octet-stream Size: 198733 bytes Desc: not available URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20180417/9d379b71/attachment.obj>