[R] lme problem

Erhart, David derhart at vividsemi.com
Fri Apr 2 00:19:55 CEST 1999


To Whom It May Concern:

Thanks in advance for any assistance.

I am trying to fit a linear mixed effects model to a reliability
degradation data set for a light emitting diode (the dependent variable
is the normalized output power, and the there are two independent
variables: the age of each device and the forward injection current). I
have linearized the model by taking logarithms of the independent and
dependent variables. There are 48 individual UNITS that are allocated
two three different CURRENT values. For each UNIT there is an NPOWER
reading at various TIMEs. The total number of observations is 332. I
have had no problem fitting models of the form,

Log(NPOWER) ~ log(TIME) + log(CURRENT), ..., random = ~ 1 + log(TIME) |
UNIT (where UNIT is the grouping variable)

I get into trouble when I try to add a random effect for log(CURRENT).
When I try to fit this model my PC runs for several minutes and then
issues a warning that the application has performed an illegal
operation. I have included the fault information at the end of this
message.

I would be very grateful for any hints. I have tried allocating more
memory ( --vsize 30M) to no avail. I do not know what is going on.

Here are some specifics about my version of the software and hardware:

> version
           _         
platform   Windows   
arch       x86       
os         Win32     
system     x86, Win32
status     Beta      
status.rev 0         
major      0         
minor      63.3      
year       1999      
month      March     
day        6         
language   R

I am using the version of lme that is in the CRAN contributed binaries
folder for WIN95/NT

Package: lme Version: 2.9-4

I am using an IBM ThinkPad 600.

I would really like to have someone try my troublesome model with my
data on a UNIX box to see if this is a PC-specific problem. If someone
can help, I would be glad to email the data set.

Regards,

David Erhart
Vivid Semiconductor
Derhart at vividsemi.com

Fault Information:

		RGUI caused an invalid page fault in
		module CRTDLL.DLL at 016f:7fd21519.
		Registers:
		EAX=00000000 CS=016f EIP=7fd21519 EFLGS=00010202
		EBX=0253dd68 SS=0177 ESP=0253daf0 EBP=0253dd50
		ECX=00000053 DS=0177 ESI=00000053 FS=0fe7
		EDX=0253dd68 ES=0177 EDI=00000053 GS=0000
		Bytes at CS:EIP:
		88 08 8b 0a 0f b6 01 41 89 0a eb 0d 52 ff 74 24 
		Stack dump:
		7fd20e08 00000053 0253dd68 0253dd38 00000003 00000000
00000000 0253df48 0000000c 0253dd84 0253dd54 02634bbc 0253df48 0394fc8c
02634bbc 0253df48

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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