Skip to content

optim hangs without warnings or error

3 messages · Katja Hebestreit, Ben Bolker, Uwe Ligges

Katja Hebestreit
# 
Hello,

optim hangs for some reason when called within the betareg function
(from the betareg package).

In this special case, the arguments which are passed to optim cause
never ending calculations.

I uploaded the arguments passed to optim on:
https://www.dropbox.com/s/ud507gbpt3gkbcp/optim_arguments.RData


I appreciate any help.
Cheers,
Katja
R version 2.15.3 (2013-03-01)
Platform: x86_64-pc-linux-gnu (64-bit)

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=C                 LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods
base     

loaded via a namespace (and not attached):
[1] betareg_3.0-2     flexmix_2.3-8     lmtest_0.9-30
modeltools_0.2-19
[5] multcomp_1.2-15   sandwich_2.2-9    stats4_2.15.3
tools_2.15.3
Ben Bolker
# 
Katja Hebestreit <katja.hebestreit <at> uni-muenster.de> writes:
Hmm, it works fine for me (thank you for the very reproducible
example!)

L <- load("/tmp/optim_arguments.RData")
arglist <- setNames(lapply(L,get),L)
do.call(optim,arglist)

Results:

$par
 mean.(Intercept) mean.groupcontrol   mean.gendermale        precision1       
precision2 
    -2.522996e+00     -5.018542e-01      6.197797e-13      4.445938e+00     
2.871008e+01 

$value
[1] 274.5367

$counts
function gradient 
     552       87 

$convergence
[1] 0

$message
NULL
R Under development (unstable) (2012-12-14 r61321)
Platform: i686-pc-linux-gnu (32-bit)

locale:
[snip]

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
[1] betareg_3.0-2     flexmix_2.3-8     lmtest_0.9-30     modeltools_0.2-19
[5] multcomp_1.2-15   sandwich_2.2-9    stats4_2.16.0     tools_2.16.0
1 day later
Uwe Ligges
# 
Katja,

this seems to be a bug.

I can reproduce this under 64-bit R-2.15.3 / R-prerelease for Windows.

It works with the results given by Ben Bolker under 32-bit R for Windows.

Will inspect shortly.

Best,
Uwe
On 07.03.2013 21:08, Katja Hebestreit wrote: