[R-SIG-Finance] [R-sig-finance] Correct specification for modelling a AR(p)-GJR GARCH(1, 1) - skewed t using fGARCH

bonjourbc9 multeesl at yahoo.co.uk
Mon Aug 31 13:56:47 CEST 2009

Hi folks, 
I am trying to model a fit for FTSE100 daily log returns. As a first step I
obtain the daily log returns using LN ( Pt / Pt-1) . Next suppose I define x
as the vector of log return series ; I used the fGARCH to model the AR(5) -
GJR GARCH(1,1) - skewed t . Can someone advise whether the following entry
is correct? 

>garchFit(FTSE100 ~arma(5,0) +aparch(1,1), data=x, init.rec = c("mci"),
delta = 2, skew = 1,
shape = 4, cond.dist = c("sstd"),
include.mean = TRUE, include.delta = FALSE, include.skew = NULL,
include.shape = NULL, leverage = NULL, trace = TRUE,
algorithm = c( "nlminb"),
control = list(), title = NULL, description = NULL)

I managed to obtain some results using this command. However the skewness
parameter returned from this command doesn't seems odd ( it is positive
value when my data exhibits negative skewness)

Can anyone help me with this code?Thanks.
View this message in context: http://www.nabble.com/Correct-specification-for-modelling-a-AR%28p%29-GJR-GARCH%281%2C1%29---skewed-t-using-fGARCH-tp25222299p25222299.html
Sent from the Rmetrics mailing list archive at Nabble.com.

More information about the R-SIG-Finance mailing list