Error Message in Matrix math- object length

Question

I am writing a code where I have an original matrix of data of 42 rows, and 6 columns. I am trying to randomly select 12 rows from this matrix (MAtrix B) and get 2 sets of regression coefficients from 2 sets of 2 columns of data within this matrix. From there, I want to use the newly calulated regression coefficients and predict the values using the data from the other 30 points in the matrix that were not selected in the random sample (Matrix C).

I keep getting an error when trying to run the code below. This is the error message I get:

Warning messages:

Warning messages:
1: 'newdata' had 30 rows but variable(s) found have 12 rows 
2: 'newdata' had 30 rows but variable(s) found have 12 rows

I am pretty sure my EDGww regression is for some reason only having 12 rows, instead of the 30 rows, which is why I am getting this error. I am not sure how to fix this though. Also, there appears to be something incorrect about "EDGww <- predict" line of code. Notsure how the "newdata" part of the command works.

A <- matrix(c(Box.Z, Box.DC.gm, Box.CR, Box.DC.ww, Box.DC.gd, Box.DC.w), nrow=42)

randco <- function(A) {

B<- A[sample(42,12),]
arows <- apply(A, 1, paste, collapse="_")
brows <- apply(B, 1, paste, collapse="_")
C<- A[-match(brows, arows), ]

Boxgm <- C[,2]
Boxww <- C[,4]
Boxgd <- C[,5]
Boxw  <- C[,6]

EDGgm<- predict(lm(B[,2] ~ B[,1]), newdata=data.frame(B=C[,1:2]))
EDGww<- predict(lm(B[,4] ~ B[,3]), newdata1=data.frame(B=C[,3:4]))


EDGgd <- EDGgm - EDGww
EDGw <- (EDGww*100/EDGgd)   



rmse.gm <- sqrt(mean((EDGgm-Boxgm)^2, na.rm = TRUE))
rmse.gd <- sqrt(mean((EDGgd-Boxgd)^2, na.rm = TRUE))
rmse.ww <- sqrt(mean((EDGww-Boxww)^2, na.rm = TRUE))
rmse.w <-  sqrt(mean((EDGw-Boxw)^2, na.rm = TRUE))

list(rmse.gm, rmse.gd, rmse.ww, rmse.w)

}

Z <- t(replicate(2, randco(A)))

Here is my matrix A:

structure(c(972.7298, 934.814227158535, 996.933797909408, 791.217703626463, 
784.234752589183, 783.659378596087, 947.066666666667, 918.351477449456, 
884.944612286002, 561.061151079137, 543.853469531525, 557.416696524543, 
545.154969718561, 549.458419120938, 572.653512815785, 508.803114948366, 
489.664296345295, 546.234253551327, 601.674754971929, 564.621528408059, 
657.593123209169, 605.194301773428, 662.548898497015, 658.731787399959, 
653.712059064807, 645.51177904143, 638.927926119253, 672.18137254902, 
662.839142363581, 679.747359701802, 696.70399323896, 673.530320189437, 
700.21186440678, 807.331830584886, 786.227683746812, 779.653232288437, 
668.38576585869, 644.662921348315, 647.371589085072, 572.055997804008, 
568.525605977766, 573.693858845096, 19.3350173135753, 19.2528485621378, 
17.97264, 19.8067268125686, 19.6961540482885, 19.2498052750475, 
18.1278467677645, 18.5052143669591, 17.8549653586694, 21.3537310834607, 
20.9012957360391, 20.7400015202543, 20.8698484361062, 20.2784363115619, 
20.4579902498884, 19.5835524259481, 20.8039105491502, 20.3545184888189, 
19.2183579751146, 19.2741515717697, 18.8531635242811, 20.4671201111593, 
19.9477324477516, 20.0816370797239, 19.16865095264, 19.3289307393948, 
19.0611214754502, 19.9355592993905, 19.8016546674182, 19.1990838235428, 
19.8209288189899, 20.0623629281521, 20.3352443372472, 19.0073567368552, 
19.0215587432765, 19.4547199391263, 18.9404044208691, 19.5176145389921, 
18.9251879854177, 20.2094551375156, 20.6700059171779, 20.6466740494858, 
0.0184126042027551, 0.0211500857551233, 0.0175062024694362, 0.0337401180409656, 
0.0346889237620637, 0.0347098429625015, 0.0197595945566906, 0.0208987649013026, 
0.0233611391296427, 0.079218078642384, 0.0854783227450312, 0.0809066783513659, 
0.0876586886055242, 0.0862439899396154, 0.0788434985488822, 0.100477074583164, 
0.109512990787781, 0.0874625785566515, 0.0697002042276112, 0.0795835405948175, 
0.0590959175682916, 0.0691508307033452, 0.0566322479499254, 0.0581804263930345, 
0.0598276441702881, 0.0615442739013907, 0.0627343073093989, 0.0496103325156931, 
0.0537150387879776, 0.0506587540839388, 0.0472721193305599, 0.0516289789530612, 
0.0470104694823737, 0.0298054466074238, 0.0322343007905799, 0.0327009202323074, 
0.0509075880344454, 0.0561708857467634, 0.055379202376338, 0.0761713062471437, 
0.0773846368854214, 0.075659018034763, 1.24742047184357, 1.25188518062349, 
1.17616931947406, 1.4929693577313, 1.44061448275529, 1.40750828224021, 
1.28417359593431, 1.35503554694099, 1.28696331263137, 2.4154456205358, 
2.34834547373797, 2.3203592861058, 2.45940290188245, 2.39036941912483, 
2.42395662840317, 2.64559953691598, 2.85551593111206, 2.78375293672456, 
2.02594247326842, 1.99500782475067, 1.96872999244972, 1.96773821995353, 
1.91969552725095, 1.93016191885846, 2.0243918375924, 2.02855863447954, 
2.0272144929465, 2.19754956724573, 2.16243045886116, 2.05659819121302, 
1.80255436993863, 1.82806964661878, 1.86868466766234, 1.59199447237071, 
1.60223256800273, 1.65077123393453, 2.34177469771875, 2.32564214914119, 
2.25709470774863, 2.31314883935834, 2.38916504448898, 2.44775420826948, 
18.0875968417318, 18.0009633815143, 16.7964698066862, 18.3137574548373, 
18.2555395655332, 17.8422969928073, 16.8436731718302, 17.1501788200181, 
16.5680020460381, 18.9382854629249, 18.5529502623012, 18.4196422341485, 
18.4104455342238, 17.8880668924371, 18.0340336214853, 16.9379528890322, 
17.9483946180381, 17.5707655520944, 17.1924155018462, 17.2791437470191, 
16.8844335318313, 18.4993818912057, 18.0280369205006, 18.1514751608654, 
17.1442591150476, 17.3003721049152, 17.0339069825037, 17.7380097321447, 
17.639224208557, 17.1424856323298, 18.0183744490513, 18.2342932815333, 
18.4665596695848, 17.4153622644845, 17.4193261752738, 17.8039487051918, 
16.5986297231503, 17.1919723898509, 16.6680932776691, 17.8963062981573, 
18.2808408726889, 18.1989198412163, 6.8965517241379, 6.95454545454546, 
7.00247928886737, 8.15217391304348, 7.89138265447493, 7.88860471724921, 
7.62407096619517, 7.90099952403618, 7.76776408558641, 12.7542993544187, 
12.6575312310825, 12.597200622084, 13.3587364700685, 13.3629275510785, 
13.441012029141, 15.6193582202551, 15.9095896423085, 15.8430941922889, 
11.783931542667, 11.5457562825984, 11.6600298656048, 10.636778199, 
10.64838914917, 10.6336366700371, 11.8079867085979, 11.7255202499558, 
11.9010541447053, 12.3889297639934, 12.2592152199763, 11.9970827762246, 
10.0039788552265, 10.0254483044328, 10.1192896841534, 9.14132274823418, 
9.19801691455233, 9.27193883373274, 14.1082410824108, 13.5274888558692, 
13.5414091470952, 12.9252863737391, 13.0692294798011, 13.4499971955802
), .Dim = c(42L, 6L), .Dimnames = list(NULL, c("col1", "col2", 
"col3", "col4", "col5", "col6")))

[Here is a link to an SO question explaining the error.](http://stackoverflow.com/questions/13965099/longer-object-length-is-not-a-multiple-of-shorter-object-length) Because your example is not reproducible, I can't easily tell what's going on. `debug` and `debugonce` may also be useful here. — Blue Magister, Dec 05 '13 at 21:17
I'm not sure. I got help with this line of code, and do not completely understand the syntax in this section. I would have no problem if it could be a matrix, but I'm, not sure how to do this. — Jason, Dec 05 '13 at 21:24
[Can you provide reproducible code?](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) — Blue Magister, Dec 05 '13 at 21:37
I added my data below. I greatly appreciate your help and if there is anyway I can e-mail you that could be helpful too. — Jason, Dec 05 '13 at 21:51
I don't know how to paste my data into my section above so it is readable. Can somebody help? I am a new at R, and appreciate all the help that has been provided thus far to me. Also, if you want to create a random matrix, with 42 rows, and 6 columns that would work to help my code be reporducible. — Jason, Dec 05 '13 at 22:16
@joran - Data has been provided the correct way as an edit above. If either of you can help, it would be greatly appreciated. Thanks again! — Jason, Dec 06 '13 at 01:29

score 3 · Accepted Answer · answered Dec 06 '13 at 01:40

You could avoid a lot of troubles if you convert A to a data.frame, allowing you to use named elements for the newdata= part of the predict call. I think this is what you are chasing, but it's hard to know for sure:

A <- setNames(data.frame(A),c("Box.Z", "Box.DC.gm", "Box.CR",
               "Box.DC.ww", "Box.DC.gd", "Box.DC.w"))


randco <- function(A) {
 rowsel <- sample(42,12)
 B <- A[rowsel,]
 C <- A[-rowsel, ]

 Boxgm <- C$Box.DC.gm
 Boxww <- C$Box.DC.ww
 Boxgd <- C$Box.DC.gd
 Boxw  <- C$Box.DC.w

 EDGgm <- predict(lm(Box.DC.gm ~ Box.Z, data=B),newdata=C[,"Box.Z",drop=FALSE])
 EDGww <- predict(lm(Box.DC.ww ~ Box.CR, data=B),newdata=C[,"Box.CR",drop=FALSE])

 EDGgd <- EDGgm - EDGww
 EDGw <- (EDGww*100/EDGgd)  

 rmse.gm <- sqrt(mean((EDGgm-Boxgm)^2, na.rm = TRUE))
 rmse.gd <- sqrt(mean((EDGgd-Boxgd)^2, na.rm = TRUE))
 rmse.ww <- sqrt(mean((EDGww-Boxww)^2, na.rm = TRUE))
 rmse.w <-  sqrt(mean((EDGw-Boxw)^2, na.rm = TRUE))

 list(gm=rmse.gm, gd=rmse.gd, ww=rmse.ww, w=rmse.w)

}

Results:

Z <- t(replicate(2, randco(A)))

#     gm        gd        ww        w       
#[1,] 0.7078684 0.7588057 0.1827982 1.310923
#[2,] 0.6728875 0.7858108 0.2072046 1.522664

Thank you so much. Your results are definitely correct. I copy and pasted your code response, and I am getting the following error "Error in randco(A) : object'A' not found". I am not quite sure why this is happening. Thank you again so much. You are a lifesaver! — Jason, Dec 06 '13 at 01:57
@Jason - works for me using your `A` matrix you posted in your question. The error sounds like you don't have an object called `A` any more. Does it appear in `ls(pattern="A")`? — thelatemail, Dec 06 '13 at 02:06
You were correct. I didn't realize I still needed my line of code with A as a matrix. Thanks you so much again!!!!! — Jason, Dec 06 '13 at 02:10

Error Message in Matrix math- object length

1 Answers1