Hello,
I am trying to do a nls regression with R. but I always get a error
as "Error in numericDeriv(form[[3L]], names(ind), env) : Missing
value or an infinity produced when evaluating the model". I googled
it and found someone said it is because of the improper start value. I
tried many times but can not solve it. Does anyone can help me?
thanks a lot !
my code is:
fit1<-nls2(lnd~log(1/(den1/R1+den2+den3+den4+den5/R5)-1)/c+log(d50),
start=c(R1=0.9, R5=23, c=-1.1, d50=10), data=SWrt)
data (SWrt) is:
Depth lnd den1 den2 den3 den4 den5
1 20 2.995732 0.4190 0.00 0.00 0.00 0.00
2 40 3.688879 0.6725 0.00 0.00 0.00 0.00
3 60 4.094345 0.8780 0.00 0.00 0.00 0.00
4 80 4.382027 0.9660 0.00 0.00 0.00 0.00
5 100 4.605170 0.9990 0.00 0.00 0.00 0.00
6 15 2.708050 0.0000 0.38 0.00 0.00 0.00
7 30 3.401197 0.0000 0.48 0.00 0.00 0.00
8 45 3.806662 0.0000 0.85 0.00 0.00 0.00
9 60 4.094345 0.0000 0.88 0.00 0.00 0.00
10 120 4.787492 0.0000 1.00 0.00 0.00 0.00
11 15 2.708050 0.0000 0.00 0.48 0.00 0.00
12 30 3.401197 0.0000 0.00 0.89 0.00 0.00
13 45 3.806662 0.0000 0.00 0.98 0.00 0.00
14 60 4.094345 0.0000 0.00 0.99 0.00 0.00
15 120 4.787492 0.0000 0.00 1.00 0.00 0.00
16 15 2.708050 0.0000 0.00 0.00 0.20 0.00
17 30 3.401197 0.0000 0.00 0.00 0.39 0.00
18 45 3.806662 0.0000 0.00 0.00 0.69 0.00
19 60 4.094345 0.0000 0.00 0.00 0.99 0.00
20 120 4.787492 0.0000 0.00 0.00 1.00 0.00
21 10 2.302585 0.0000 0.00 0.00 0.00 139.50
22 30 3.401197 0.0000 0.00 0.00 0.00 227.85
23 50 3.912023 0.0000 0.00 0.00 0.00 306.90
24 70 4.248495 0.0000 0.00 0.00 0.00 381.30
25 90 4.499810 0.0000 0.00 0.00 0.00 432.45
26 110 4.700480 0.0000 0.00 0.00 0.00 455.70
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.
