Hello,
I am fitting curves in drc to a Log-logistic 4-parameter model.
My experiment is looking at fall and spring applications of a herbicide on
wild oat at 5 different locations. A preliminary mixed model ANOVA showed
that timing was significant. I am running a dose response curve for each
application timing at each location. However, when I run them all
together, my one fall location has estimated b,c,d and e, however the
standard error for all of them is NaNS. When I run only the fall location
on its own, there is no problem in getting standard errors for all my
parameters.
What is the difference between running it alone and running it with other
curves? How can I fix the NaNS standard errors in the multiple curves as I
need to be able to compare ED50s.
The data for the curve alone is below. When I run it with the rest of the
curves they are separated by biotype. Dependent variable is square root
biomass (srbiomass) and treatment (Trt) is the independent variable.
Location Timing biotypeno Trt Rep Trtno biomass srbiomass Kinsella Fall 5
0 1 1 21.2 4.604346 Kinsella Fall 5 0 2 1 168.72 12.98923 Kinsella Fall 5
0 3 1 86.76 9.314505 Kinsella Fall 5 0 4 1 14.3 3.781534 Kinsella Fall 5
50 1 2 12.7 3.563706 Kinsella Fall 5 50 2 2 46.44 6.81469 Kinsella Fall 5
50 3 2 9.84 3.136877 Kinsella Fall 5 50 4 2 32.38 5.690343 Kinsella Fall 5
100 1 3 10.3 3.209361 Kinsella Fall 5 100 2 3 41.84 6.468385 Kinsella Fall
5 100 3 3 0 0 Kinsella Fall 5 100 4 3 4.32 2.078461 Kinsella Fall 5 150 1
4 3.84 1.959592 Kinsella Fall 5 150 2 4 9.34 3.056141 Kinsella Fall 5 150
3 4 25.9 5.089204 Kinsella Fall 5 150 4 4 1.46 1.208305 Kinsella Fall 5
200 1 5 6.6 2.569047 Kinsella Fall 5 200 2 5 0 0 Kinsella Fall 5 200 3 5
72.94 8.540492 Kinsella Fall 5 200 4 5 24.56 4.955805 Kinsella Fall 5 300
1 6 6.36 2.521904 Kinsella Fall 5 300 2 6 3.6 1.897367 Kinsella Fall 5 300
3 6 31.28 5.592853 Kinsella Fall 5 300 4 6 1.46 1.208305 Kinsella Fall 5
400 1 7 2.02 1.421267 Kinsella Fall 5 400 2 7 11.2 3.34664 Kinsella Fall 5
400 3 7 3.02 1.737815 Kinsella Fall 5 400 4 7 0 0
Thanks in advance for any help.
Breanne
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
