Skip to content

adding columns to a table after a loop

5 messages · Albert Picado, Uwe Ligges, John Kane

Original
Albert Picado
# 
Dear list,

I am trying to find the way to add columns to a table
(or a matrix) after a loop. The result of this loop is
an array and I would like to get all the results in a
single table once the loop is finish. I have
reproduced a simplified example below:
+ b<-rnorm(10)
+ a<- a +1
+ }

# I have tried several methods but without succes so
far.
b
1  -0.03250661
2  -0.59823770
3   1.58120471
4   0.41086546
5   0.78959090
6   1.23587125
7   0.83427190
8   1.09035581
9   0.11331056
10  0.25267231

the result above is not what I am looking for because
it shows only the results from the last operation in
the loop, I would like to obtain something like the
following.
b1         b2          b3         b4
1  -0.08420826 -0.0637943 -0.83246201 -1.0902384
2   0.40623009 -2.7940096  1.37664973  1.6023967
3  -1.13850505  1.1660669 -0.95962296 -0.7325098
4  -1.06183391  1.1063677  1.67948677 -1.9875475
5  -0.64431067 -0.4843952 -1.30742858  0.5064134
6   0.56729468  1.0860484  0.07651954  0.4380108
7   0.95036177 -0.5328609  1.28954934 -0.2775614
8  -0.17848223  0.5340379 -0.22613700  1.0179886
9   2.93145454 -1.8639607 -0.25478610  1.6619754
10  1.51942415 -0.2051423  0.09144450 -1.6329481

I hope someone can give me a hand

best regards

a




	
 p5.vert.ukl.yahoo.com uncompressed/chunked Sat Aug 12 13:14:00 GMT 2006
Uwe Ligges
#
Albert Picado wrote:
Staying in your example:

a <- 1
b <- matrix(nrow=10, ncol=4)
while(a <= 4){
     b[,i] <- rnorm(10)
     a <- a + 1
}

Uwe Ligges
Albert Picado
# 
Dear Uwe,
 
Thanks for your answer. I have been trying your suggestion but I haven't been able to solve my problem as I may have simplified in excess my example. The real situation is actually much similar to the following:
 
a<-1
b<-1
i<-1
pvalue<-matrix()
while(a<=4) {
     while(b<=10){
     pvalue[i]<-rnorm(1)
     i<-i+1
     b<-b+1
     }
a<-a+1
}

To get the desired outcome I have tried two things based on your suggestion:
 
1)
 
a<-1
b<-1
i<-1
pvalue<- matrix(ncol=4, nrow=10)
result<- matrix()
while(a<=4) {
 while(b<=10){
 pvalue[i]<-rnorm(1)
 i<-i+1
 b<-b+1
 }
result[,a]<-pvalue
a<-a+1
}

here I get an error message: 
Error in "[<-"(`*tmp*`, , a, value = c(1.38661984066808, -0.641565430307799,  : 
        number of items to replace is not a multiple of replacement length

2)
 
a<-1
b<-1
i<-1
pvalue<-matrix(ncol=4, nrow=10)
while(a<=4) {
 while(b<=10){
 pvalue[i,a]<-rnorm(1)
 i<-i+1
 b<-b+1
 }
a<-a+1
}

here what I get is:
[,1] [,2] [,3] [,4]
 [1,] -2.18379100   NA   NA   NA
 [2,] -0.90112392   NA   NA   NA
 [3,] -0.91762163   NA   NA   NA
 [4,]  0.54922398   NA   NA   NA
 [5,]  0.95511590   NA   NA   NA
 [6,]  0.08509518   NA   NA   NA
 [7,] -0.92576835   NA   NA   NA
 [8,]  1.62361191   NA   NA   NA
 [9,]  0.02644486   NA   NA   NA
[10,]  0.32406407   NA   NA   NA

 
I do not know what else to try...
 
best wishes
 
a


----- Message d'origine ----
De : Uwe Ligges <ligges at statistik.uni-dortmund.de>
? : Albert Picado <albert_picado at yahoo.fr>
Cc : r-help at stat.math.ethz.ch
Envoy? le : Samedi, 12 Ao?t 2006, 3h57mn 16s
Objet : Re: [R] adding columns to a table after a loop
Albert Picado wrote:
Staying in your example:

a <- 1
b <- matrix(nrow=10, ncol=4)
while(a <= 4){
     b[,i] <- rnorm(10)
     a <- a + 1
}

Uwe Ligges
Uwe Ligges
#
Albert Picado wrote:
I think you will have to read some basic documentation such as "An 
Introduction to R" about data structures, indexing, and loops in R 
before proceeding ...

Anyway, you can do:

a <- 1
pvalue <- matrix(nrow = 10, ncol = 4)
while(a <= 4) {
     b <- 1
     while(b <= 10){
         pvalue[b, a] <- rnorm(1)
         b <- b+1
     }
     a <- a+1
}


or much shorter


pvalue <- matrix(nrow = 10, ncol = 4)
for(a in 1:4) {
     for(b in 1:10){
         pvalue[b, a] <- rnorm(1)
     }
}


Uwe Ligges
John Kane
#
--- Albert Picado <albert_picado at yahoo.fr> wrote:
Albert
----clip
There is a minor typo in the above  example. b[,1]
should read b[,a].  It works very nicely with that
small correction.