# [R] Addressing results of modFit() of the FME package

Bachelor student d.wintermantel at gmx.de
Thu Jan 31 10:12:42 CET 2013

```Hello,

I don´t manage to address the results I get from modFit(). Would be really
nice if you could help me to do so.

This is my loop:

for (i in 1:10) {

CO2 <- CO2.list[[i]]

#Beobachtungen

Obs <- data.frame("x=t"=numeric(),"y=Respi"=numeric())
Obs <- cbind(CO2[,1],CO2[,3])
colnames(Obs) <- c("x","y")

Corg.mg <- CO2\$Corg.mg

Model<-function(p,x) return(data.frame(x=x,
y=(p[1]*p[2]*exp((-p[2])*x))+((Corg.mg-p[1])*p[3]*exp((-p[3]*x)))))
Residuals<-function(p)(Obs[,2]-Model(p, Obs[,1])\$y)

print(system.time(P <- modFit(f=Residuals, p=c(300, 0.1, 0.002),
lower=c(0,0,0), upper=c(500, 5, 1))))
sP<-summary(P)
sP

}

#sP, gives me the following output. I can address the estimates with P\$par
but I cannot address Std. Error and Pr(>t). That´s what I would like to do!

Parameters:
Estimate Std. Error t value Pr(>|t|)
[1,] 4.005e+00 2.510e+00 1.595 0.131469
[2,] 1.603e-01 9.442e-02 1.698 0.110100
[3,] 3.647e-04 8.581e-05 4.251 0.000698 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.2427 on 15 degrees of freedom

Parameter correlation:
[,1] [,2] [,3]
[1,] 1.0000 -0.9164 -0.7697
[2,] -0.9164 1.0000 0.5599
[3,] -0.7697 0.5599 1.0000