R help - Mar 2012 - How do I force confint() for glm() to be quiet?

I need confint() for glm() to supress the messages 

"Waiting for profiling to be done..."

because they mess up the caching mechanism of pgfSweave (see
https://github.com/cameronbracken/pgfSweave/issues/40).

I have read the help page of confint(), but I do not know how to get
the help page for the glm() version, if any such help page exists.

Is there a general way of turning of output from functions in R, that
would help here?

Below is an example of an intended usage scenario:

x <- 10000
set.seed(42)
a <- rnorm(x)
b <- factor(LETTERS[sample(1:7, x, replace = TRUE)])
c <- factor(LETTERS[sample(1:4, x, replace = TRUE)])
my.fit <- glm(c ~ b + a, family = "binomial")
my.results <- confint(my.fit)

In stats:::confint.glm (actually it is MASS:::confint.glm) you see that the
message() function is used to produce "waiting for ...". To avoid this
message you can just use 
suppressMessages(confint(my.fit))
At least in R it works, hope it works with Sweave as well. 

Regards,
Sina

--
View this message in context:
http://r.789695.n4.nabble.com/How-do-I-force-confint-for-glm-to-be-quiet-tp4459152p4459227.html
Sent from the R help mailing list archive at Nabble.com.

On Mar 9, 2012, at 6:14 AM, Hans Ekbrand wrote:
> I need confint() for glm() to supress the messages
I'm wondering if suppressMessages would be helpful? Which in turn  
suggests that you do not know how to use "??", so firt you should get
in the habit of doing a helpSearch before posting.

??"suppress messages"
>
> "Waiting for profiling to be done..."
>
> because they mess up the caching mechanism of pgfSweave (see
> https://github.com/cameronbracken/pgfSweave/issues/40).
>
> I have read the help page of confint(), but I do not know how to get
> the help page for the glm() version, if any such help page exists.
When I type ?confint.glm at my console I get this help page:

confint-MASS {MASS}

>
> Is there a general way of turning of output from functions in R, that
> would help here?
If suppressMessages is not effective then look at:

?sink

>
> Below is an example of an intended usage scenario:
>
> x <- 10000
> set.seed(42)
> a <- rnorm(x)
> b <- factor(LETTERS[sample(1:7, x, replace = TRUE)])
> c <- factor(LETTERS[sample(1:4, x, replace = TRUE)])
> my.fit <- glm(c ~ b + a, family = "binomial")
> my.results <- confint(my.fit)
>
Grrrr. A _minimal_ example would have had fewer iterations, but this  
does seem to be effective:

suppressMessages(my.results <- confint(my.fit))


-- 
David Winsemius, MD
West Hartford, CT