[R] Capturing coxph warnings and errors

[Stephen Weigand]

On Tue, Jun 17, 2008 at 9:47 AM, Daniel Brewer <daniel.brewer at icr.ac.uk> wrote:
> Hi,
>
> I have a script that takes a subset of genes on a microarray and tries
> to fit a coxph model to the expression values for each gene.  This seems
> to work fine but in some cases it produces warnings and/or errors.
>
> For example:
> Error in fitter(X, Y, strats, offset, init, control, weights = weights,  :
>  NA/NaN/Inf in foreign function call (arg 6)
> In addition: Warning message:
> In fitter(X, Y, strats, offset, init, control, weights = weights,  :
>  Ran out of iterations and did not converge
>
> In this situation I would like to:
> 1) Deal with it so that it the scripts result data.frame the p value etc
> are set to NA.
> 2) Suppress the Error and Warning messages
>
> What is the best way to do this?
>

Perhaps look at try() and look at the help for options() which
explains different ways to handle warnings and errors.