[R] why does arima returns "NAN" standard error?

Spencer Graves spencer.graves at pdf.com
Sun Jun 4 19:35:06 CEST 2006


	  Your question does not include a simple, self-contained example (as 
requested in the posting guide! "www.R-project.org/posting-guide.html"). 
  Without that, I don't even know where to start.  However, the model 
appears to be overparameterized and NOT invertible.  Have you made "acf" 
and "pacf" plots?

	  Beyond that, have you studied the time series chapter in Venables and 
Ripley (2002) Modern Applied Statistics with S (Springer)?  If no, I 
suggest you do that before you do much else.  After doing that, if you 
still have a question for this list, please submit another post, 
preferably including a simple, self-contained example.

	  Hope this helps,  	
	  Spencer Graves	

Michael wrote:
> Hi everyone,
> 
> 
> -----------------------------
> 
> Coefficients:
> 
>          ar1      ar2      ma1     ma2    sar1  intercept   drift
> 
>       1.5283  -0.7189  -1.9971  0.9999  0.3982     0.0288  -9e-04
> 
> s.e.  0.0869   0.0835   0.0627  0.0627  0.1305        NaN     NaN
> 
> 
> 
> sigma^2 estimated as 0.04383:  log likelihood = 4.34,  aic = 7.32
> 
> Warning message:
> 
> NaNs produced in: sqrt(diag(object$var.coef))
> 
>  -------------------------------------------
> 
> 
> What does this mean?
> 
> 
> Thanks a lot!
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html



More information about the R-help mailing list