R help - Aug 2006 - Protection stack overflow

G'day all.

I'm a new user of R -- but an arms-length user, as I'm running it from
Octave via the ROctave interface that Duncan Temple Lang wrote some
years ago and Stephan Van Der Walt recently updated for use with Octave
2.1.71. I'm using R version 2.1.1. ROctave uses libR.so to provide the
interface. My system is Ubuntu linux 5.10 and I'm using the packages
that come with this distro.

I'm getting a protection stack overflow error when recursively
calculating AR(p) time series models using the arima() function. The
recursion is involved because I calculate a new model with each new time
series data point. (I'm trying to reproduce some results in a research
paper and this is what they're doing.)

I've tried setting the expressions parameter to a higher number using
options(expressions=500000) but I'm still getting this problem with
stack overflow. I can get about 400-odd iterations and then the overflow
error appears. I yet haven't tried running the recursion natively in R,
and I realise I should do that.

Your advice would be much appreciated.

Cheers,
Paul.

R has a command-line option to set the ppstack size,

 --max-ppsize=N        Set max size of protect stack to N

Looks like you need to supply this (and it can be done with embedded R)
if the problem persists with current R.

You could also try arima0 or even ar to do the fitting.

On Tue, 15 Aug 2006, Paul Koufalas wrote:
> G'day all.
> 
> I'm a new user of R -- but an arms-length user, as I'm running it
from
> Octave via the ROctave interface that Duncan Temple Lang wrote some
> years ago and Stephan Van Der Walt recently updated for use with Octave
> 2.1.71. I'm using R version 2.1.1. ROctave uses libR.so to provide the
> interface. My system is Ubuntu linux 5.10 and I'm using the packages
> that come with this distro.
Note that your version of R is well outdated, and the posting guide did 
ask you to update it BEFORE posting.
> I'm getting a protection stack overflow error when recursively
> calculating AR(p) time series models using the arima() function. The
> recursion is involved because I calculate a new model with each new time
> series data point. (I'm trying to reproduce some results in a research
> paper and this is what they're doing.)
> 
> I've tried setting the expressions parameter to a higher number using
> options(expressions=500000) but I'm still getting this problem with
> stack overflow. I can get about 400-odd iterations and then the overflow
> error appears. I yet haven't tried running the recursion natively in R,
> and I realise I should do that.
> 
> Your advice would be much appreciated.
> 
> Cheers,
> Paul.
> 
> ______________________________________________
> 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
> and provide commented, minimal, self-contained, reproducible code.
> 
-- 
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

Thanks for your reply, Prof Ripley.

I tried the --max-ppsize option and now ROctave (R embedded in Octave)
no longer produces the protection stack overflow error I was seeing with
the default.

Specifically, in the file RinOctave.cc, in the function initialize(), I
put in the --max-ppsize option as follows:

        const char * const DefaultArgs[]
{"ROctave","--max-ppsize=100000"};

where 100000 is the maximum I found I can use with R 2.1.1.

Cheers,
Paul.


Prof Brian Ripley wrote:
>R has a command-line option to set the ppstack size,
>
> --max-ppsize=N        Set max size of protect stack to N
>
>Looks like you need to supply this (and it can be done with embedded R)
>if the problem persists with current R.
>
>  
>