[R] Problems using plsr() in the pls package

[CG Pettersson]

Dear all,
I am trying to do a PLS regression using plsr() from the pls package.
I have tried to make a dataset according to the methods recommended in "Introduction to the pls Package" (Mevik & Wehrens, 2015) and it looks like I get a structure in the dataset that resembles what is packed with the package:

> str(gasoline)
'data.frame':   60 obs. of  2 variables:
$ octane: num  85.3 85.2 88.5 83.4 87.9 ...
$ NIR   : AsIs [1:60, 1:401] -0.0502 -0.0442 -0.0469 -0.0467 -0.0509 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr  "1" "2" "3" "4" ...
  .. ..$ : chr  "900 nm" "902 nm" "904 nm" "906 nm" ...

> str(lid_allt2)
'data.frame':   223 obs. of  2 variables:
$ DON     : num  55 55 55 55 55 55 55 55 55 55 ...
$ lid_nose: 'AsIs' chr [1:223, 1:192] "215.074835" "203.803736" "197.364614" "192.765756" ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr  "1" "2" "3" "4" ...
  .. ..$ : chr  "n1" "n2" "n3" "n4" ...

Where "gasoline" is a dataset from the package and "lid_allt2" is my own dataset. The reason there are so many "55" in the example row in the str() call is that I have repeated measurments with an electronic nose, normally six repetition for each control measurement.

When I try to explain the DON variable with "lid_nose", the plsr() accepts my coding but returns a error message after on second or so:

> don1 <- plsr(DON ~ lid_nose, ncomp = 10, data = lid_allt2, validation = "LOO")
Error in `[[<-.data.frame`(`*tmp*`, i, value = c(16088L, 14624L, 14080L,  :
  replacement has 42816 rows, data has 223

I have clearly asked plsr() to do something very stupid, but what, and what should I do to come out a little better?

Med vänlig hälsning/Best regards
CG Pettersson
Senior Scientific Advisor, PhD
______________________
Lantmännen Corporate R&D
Phone:  +46 10 556 19 85
Mobile: + 46 70 330 66 85
Email: cg.pettersson at lantmannen.com<mailto:cg.pettersson at lantmannen.com>
Visiting Address: S:t Göransgatan 160 A
Address: Box 30192, SE-104 25 Stockholm
Webb: http://www.lantmannen.com<http://www.lantmannen.com/>
Registered Office: Stockholm
Before printing, think about the environment


	[[alternative HTML version deleted]]