0

This might be a long shot, but here goes. I performed experiments based on a Box Behnken design using R, I plugged in my HPLC data and the 3 conditions at 3 levels.

I am trying to produce response surface and contour plots from my code using R-studio. My code (more like one that's being used in the lab) is able to perform all the statistical calculations, but it gives me an error when I need it to generate the plots. Please refer to the image for the code, the next chunk (not shown in the image), this is the code is the one that gives me an error message screengrab of the code


pdf("HPLC_BB1_PR.pdf" ,paper="a4r")
for (i in 1:length(myvar)){
  par(mfrow=c(2,2),cex.axis=0.4,cex.lab=0.65,cex.sub=0.7)
  persp(mymodel[[i]],~x1+x2+x3,at=summary(mymodel[[i]])$canonical$xs,zlab=myvar[i],col=rainbow(100),contours="colors")
  print(i)
}[enter image description here][1]
dev.off()

the Error message I get is: Error in contour.lm(x, form, at, bounds, zlim, xlabs, atpos = atpos, plot.it = FALSE) : 'at' must be a NAMED list of values used for surface slices

Any help would be greatly appreciated. I've been stuck for like a week now.

BBD values and Error message in the console

Phil
  • 7,287
  • 3
  • 36
  • 66
Viwe Tyibilika
  • 1
  • 1
  • What does `summary(mymodel[[i]])$canonical$xs` look like? From the error message it sounds like it is unnamed, while the function is expecting it to be named. – jpiversen Mar 20 '22 at 15:42
  • You should put together a small self-contained example. See https://stackoverflow.com/q/5963269/2554330 . – user2554330 Mar 20 '22 at 16:40

0 Answers0