[R] R-package rmgarch
Stefanos charalampopoulos
st.charalampopoulos at gmail.com
Tue Mar 8 12:58:29 CET 2016
Hi all,
The following questions is regarding to the statistical software R.
Specifically refers to the rmgarch package for time series analysis.
I have a dataset of [1520,34]. I am trying to fit a dcc. I am using
multispec because i want to include many univariate (c(spec1,spec2,spec3)
etc)models to have the comparison as it is done in Cappiello et all (2006).
When after that i am using multifit the model collapses resulting in
error :speclist length not equal to data length.
code: spec1 = ugarchspec(mean.model = list(armaOrder = c(1, 1),include.mean
= TRUE), variance.model = list(garchOrder = c(1,1), model = 'eGARCH'),
distribution.model = 'norm')
spec2 = ugarchspec(mean.model = list(armaOrder = c(1, 1),include.mean =
TRUE), variance.model = list(garchOrder = c(1,1), model = 'sGARCH'),
distribution.model = 'norm')
spec3 = ugarchspec(mean.model = list(armaOrder = c(1, 1),include.mean =
TRUE), variance.model = list(garchOrder = c(1,1), model = 'sGARCH'),
distribution.model = 'norm')
spec4 = ugarchspec(mean.model = list(armaOrder = c(1, 1),include.mean =
TRUE), variance.model = list(garchOrder = c(1,1), model = 'gjrGARCH'),
distribution.model = 'norm')
mspec = multispec( replicate(8.5,c( spec1, spec2, spec3, spec4 )))###i am
using 8.5 to have 8.5x4=34.
cl = makePSOCKcluster(34)
multf = multifit(mspec, data1, cluster = cl,)
Any suggestions?
Thanks in advance,
Best,
Stefanos
[[alternative HTML version deleted]]
More information about the R-help
mailing list