R help - Oct 2005 - arima: warning when fixing MA parameters.

I am puzzled by the warning message in the output below.  It appears
whether or not I fit the seasonal term (but the precise point of doing
this was to fit what is effectively a second seasonal term).  Is there
some deep reason why AR parameters
("Warning message: some AR parameters were fixed: ...")
should somehow intrude into the fitting of a model that has only MA  
terms?

 > library(DAAG)
 > attach(bomsoi)
 > # The following is fine:
 > arima(avrain, order=c(0,0,4), seasonal=list(order=c(0,0,1),  
period=12),
+  fixed=c(NA,0,0,NA,NA,NA))
.....
 > # The following generates a warning message
 > arima(avrain, order=c(0,0,4), seasonal=list(order=c(0,0,1),  
period=12),
+  fixed=c(0,0,0,NA,NA,NA))

Call:
arima(x = avrain, order = c(0, 0, 4), seasonal = list(order = c(0, 0,  
1), period = 12),
     fixed = c(0, 0, 0, NA, NA, NA))

Coefficients:
       ma1  ma2  ma3     ma4     sma1  intercept
         0    0    0  0.0357  -0.1061   456.6675
s.e.    0    0    0  0.1015   0.0886     7.6997

sigma^2 estimated as 6849:  log likelihood = -595.23,  aic = 1198.46
Warning message:
some AR parameters were fixed: setting transform.pars = FALSE in:  
arima(avrain, order = c(0, 0, 4), seasonal = list(order = c(0,


John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.

It's a bug: the code has 1:arma[1], i.e. 1:0.  Replace by 
seq(length=arma[1]).

On Thu, 13 Oct 2005, John Maindonald wrote:
> I am puzzled by the warning message in the output below.  It appears
> whether or not I fit the seasonal term (but the precise point of doing
> this was to fit what is effectively a second seasonal term).  Is there
> some deep reason why AR parameters
> ("Warning message: some AR parameters were fixed: ...")
> should somehow intrude into the fitting of a model that has only MA
> terms?
>
> > library(DAAG)
> > attach(bomsoi)
> > # The following is fine:
> > arima(avrain, order=c(0,0,4), seasonal=list(order=c(0,0,1),
> period=12),
> +  fixed=c(NA,0,0,NA,NA,NA))
> .....
> > # The following generates a warning message
> > arima(avrain, order=c(0,0,4), seasonal=list(order=c(0,0,1),
> period=12),
> +  fixed=c(0,0,0,NA,NA,NA))
>
> Call:
> arima(x = avrain, order = c(0, 0, 4), seasonal = list(order = c(0, 0,
> 1), period = 12),
>     fixed = c(0, 0, 0, NA, NA, NA))
>
> Coefficients:
>       ma1  ma2  ma3     ma4     sma1  intercept
>         0    0    0  0.0357  -0.1061   456.6675
> s.e.    0    0    0  0.1015   0.0886     7.6997
>
> sigma^2 estimated as 6849:  log likelihood = -595.23,  aic = 1198.46
> Warning message:
> some AR parameters were fixed: setting transform.pars = FALSE in:
> arima(avrain, order = c(0, 0, 4), seasonal = list(order = c(0,
>
>
> John Maindonald             email: john.maindonald at anu.edu.au
> phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
> Centre for Bioinformation Science, Room 1194,
> John Dedman Mathematical Sciences Building (Building 27)
> Australian National University, Canberra ACT 0200.
>
> ______________________________________________
> 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
>
-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595