0

I am trying to test R for non linear constrained optimization. I have 2 files: T1.csv has

 DV  LB  UB  Hist_Inv
X1  0.7 1.3     28462739.43
X2  0.7 1.3      177407.18
X3  0.7 1.3      1423271.12

and T2.csv has 

Count  DV  weight_f    weight_v  p  coef
1   X1  2.310281831 3.661156016 0.5 1828.105881
2   X1  0.693084549 2.20503016  0.5 1460.686147
3   X1  0.207925365 2.030522789 0.5 1436.277144
4   X1  0   5.248353307 0.8 1050.493355
5   X1  0   1.591805116 0.8 983.9964128
6   X1  0   1.933056056 0.8 459.9371809
7   X2  7.322516444 138 0.5 387.4659072
8   X2  3.661258222 139 0.5 606.8684771
9   X2  1.830629111 176.5   0.5 358.8902965
10  X3  164294.4758 77024   0.2 282.0477107
11  X3  98576.68545 122261.4    0.2 345.9217482
12  X3  59146.01127 166242.84   0.2 364.9587162

And the code is...

df <- read.csv("C:/Users/prashant.mudgal/Downloads/T1.csv")
dv <- read.csv("C:/Users/prashant.mudgal/Downloads/T2.csv")
decVars <- dv$DV
coeff <- dv$coef
wf<-dv$weight_f
wv<-dv$weight_v
p<-dv$p
y<-0
inv<-df$Hist_Inv

fr <- function(decVars) {  
    for(i in 1:length(decVars)){
        k <-coeff[i]*(wf[i] + wv[i] * decVars[i])**p[i]
        y <- y + k
    }
    y
}

temp<-0
for(j in 1:length(inv)){
    temp[j]<- 1
}


optim(temp, fr, NULL,control=list(fnscale=-1))
##constrOptim(inv,fr,NULL,Control=list(fnscale=-1))

I have to use constraints in optimization. Cons are in my T1.csv file, specified as UB and LB such that 

0.7<=X1<=1.3
0.7<=X2<=1.3
0.7<=X3<=1.3

How do I specify the cons in my code? (I don't want to hard code them, I want to read them from the file T1.csv.)

Ben Bolker
  • 192,494
  • 24
  • 350
  • 426
prashant mudgal
  • 21
  • 6
  • Use constroptim : http://stackoverflow.com/questions/16345271/setting-constraints-in-constroptim Then read your csv, put it in data.frame and mention the constraints value contained in it. – Colonel Beauvel Sep 15 '14 at 17:28
  • you don't need constrOptim, you just need box constraints (i.e. see `optim`, try `method="L-BFGS-B"`, and look at the `lower` and `upper` arguments. `lower` and `upper` can be specified as vectors that are columns of `df`. – Ben Bolker Sep 15 '14 at 17:40

0 Answers0