[R] nls error handling

Gregg Rosenkranz gregg_rosenkranz at fishgame.state.ak.us
Thu Jan 11 03:05:17 CET 2001

I'm written a function that uses nls in a loop to fit a model to resampled data.  Occasionally a sample is drawn such that the algorithm (apparently) can't converge and 

step factor reduced below minimum

is printed to the screen.  I tried a few different settings for the minFactor parameter in nls.control with no different result.  When this happens, I would like my function to put NAs in the output matrix that holds the parameter estimates and move to the next iteration.  As it is, the function terminates upon error and the partially-filled result matrix is not returned.  Any suggestions?

platform i386-pc-mingw32
arch     x86            
os       Win32          
system   x86, Win32     
major    1              
minor    2.0            
year     2000           
month    12             
day      15 

Thanks for your time,

Gregg Rosenkranz
Scallop Biometrician
Alaska Department of Fish and Game 
211 Mission Road
Kodiak, AK 99615

phone 907.486.1858
fax 907.486.1824
gregg_rosenkranz at fishgame.state.ak.us
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list