Hi All,
I'm trying to use non linear regression to my data in order to to
prediction.
First of all the model is very close to "linear" but not so good in
prediction so I try to fit non linear function to my data.
Here is my datas :
datalm15
Y M V X
1 512 485 8 1
2 1029 811 14 2
3 1566 1040 17 3
4 1967 1178 20 4
5 2364 1272 23 5
6 2720 1310 24 6
7 3096 1457 26 7
8 3420 1514 28 8
9 3797 2504 37 9
10 4185 2533 38 10
11 4649 3065 46 11
12 5083 3179 50 12
13 5455 3261 51 13
14 5883 3640 60 14
15 6407 3680 61 15
and my model (which is convergent with this example) :
csreg.st<-c(b1= 30000, b2= 4, b3= -2)
csreg.md<- Y ~ b1/(1+exp(b2+b3*X))
csreg.fm <- nls(csreg.md, start=csreg.st, trace=TRUE)
Now I want to see with my data set limited to the first 15 values :
datalm<-datalm[1:15,]
And with the same code I received the following error :
Erreur dans qr.qty(QR, resid) :
'qr' and 'y' must have the same number of rows
And :
In .swts * attr(rhs, "gradient") :
longer object length is not a multiple of shorter object length
I do not understand really what's going on here.
if somebody have a hint it could be very helpfull :)
All the best,
Thomas
[[alternative HTML version deleted]]
_______________________________________________
R-SIG-Mac mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-mac
