Hi all,

I want to fit the following model to my data:


Y_t= a+bY_(t-1)+cY_(t-2) + Z_t +Z_(t-1) + Z_(t-2) + X_t + M_t

i.e. it is an ARMA(2,2) with some additional regressors X and M.

[Z_t's are the white noise variables]




So, I run the following code:


for (i in 1:rep)  {            index=sample(4,15,replace=T)

 final<-do.call(rbind,lapply(index,function(i)
get(as.character(levels[as.numeric(levels)==i]))))
 --------------------------------> this gives me my data set


                                 l=final$l
                                 g=final$g



 model=arima0(l,order=c(2,0,2),xreg=g,method=("ML"))

                          }



I get warnings like:

Call:
arima0(x = l, order = c(2, 0, 2), xreg = g, method = ("ML"))

Coefficients:
         ar1      ar2      ma1     ma2  intercept  xreg
      0.2549  -0.7722  -0.5586  0.3563      3e-04     0
s.e.  0.1072   0.0866   1.6725  0.2221      2e-04   NaN

sigma^2 estimated as 3.921e-06:  log likelihood = 311.72,  aic = -609.44



Warning message:
In sqrt(diag(x$var.coef)) : NaNs produced


What does this output mean?
How do I get rid of this and do my analysis properly?

Any help is welcome.

Thanks,
Preetam




-- 
Preetam Pal
(+91)-9432212774
M-Stat 2nd Year,                                             Room No. N-114
Statistics Division,                                           C.V.Raman
Hall
Indian Statistical Institute,                                 B.H.O.S.
Kolkata.

        [[alternative HTML version deleted]]

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to