[R] Functions ,Optim, & Dataframe

[Dieter Menne]

Michael Papenfus <mmpapenf <at> wisc.edu> writes:

> 
> I have defined the following function:
> 
> fr<-function(x) {
>     u<-x[1]
>     v<-x[2]
>     sqrt(sum((plnorm(c(3,6),u,v)-c(.55,.85))^2))
> }
> which I then solve using optim
> y<-optim(c(1,1),fr)
> 
>  > y$par
> [1] 1.0029771 0.7610545
> This works fine.
> 
> Now I want to use these two steps on a dataframe:
>  mydat<-data.frame(d1=c(3,5),d2=c(6,10),p1=c(.55,.05),p2=c(.85,.35))
>  > mydat
>   d1 d2   p1   p2
> 1  3  6 0.55 0.85
> 2  5 10 0.05 0.35
> 
> where for each row in mydat, I append the two parameter resulting from 
> optim into mydat.
> I want to do this for a larger dataset but thought I would start with a 
> simple two row dataframe.
> 

I would prefer a loop in this case.

fr<-function(x) {
    sqrt(sum((plnorm(c(3,6),x[1],x[2])-c(x[3],x[4]))^2))
}
y<-optim(c(1,2,0.55,0.85),fr)

mydat<-data.frame(d1=c(1,0.5),d2=c(1,0.1),p1=c(.55,.05),p2=c(.85,.35))
myres<-mydat   # simple way to allocate dataframe for results
names(myres) = paste("res",names(myres),sep=".")

for (i in 1:nrow(mydat)){
  y <- optim(mydat[i,1:4],fr)
  myres[i,] <- y$par
}
mydat = cbind(mydat,myres)