R help - Jan 2005 - Please use colMeans()! was: Re: Calculate Mean of Column Vectors?

There are indeed speed advantages in using colSums etc. However the disadvantage
is that the newbie doesn't always find the power inherent in the apply,
sapply, tapply and mapply. For many things that I do, the speed is the least of
my worries; although I take the point that using apply for means or sums in
packages that are distibuted to others is not the way to go.

As many of us have found out (and I think it was in S Poetry) the statement was
made that vectorisation befuddles some beginners. So learning how to use this
command, on a nice easy topic such as summing or averaging does have some merit.
I have to admit using colSums a lot, but I don't think I have ever thought
to use colMeans.

As an aside; does anyone have a list of optimised functions. That is functions
like this one whose main benefit is speed (I think) I guess many of us feel that
we are constantly using hammers to crack the nuts, but we still don't really
know what's in the toolbox.

Tom
> -----Original Message-----
> From: Uwe Ligges [mailto:ligges at statistik.uni-dortmund.de]
> Sent: Tuesday, 11 January 2005 3:57 PM
> To: r-help at stat.math.ethz.ch
> Cc: 'Thomas Hopper'
> Subject: Please use colMeans()! was: Re: [R] Calculate Mean of Column
> Vectors?
> 
> 
> Folks, please see ?colMeans (as already pointed out in at least one 
> message) and friends.
> These are faster than those apply(X, 2, mean) calls
> and please do use them when publishing code (both in packages 
> in on this 
> list).
> 
> Uwe Ligges
> 
> 
> >>Hello,
> >>
> >>I've got an array defined as y <- rnorm(3000), dim(y) <-
c(3, 1000).
> >>
> >>I'd like to produce a 1000-element vector z that is the mean of
the
> >>corresponding elements of y (like z[1,1] <- mean(y[1,1], y[2,1],
> >>y[3,1])), but being new to R, I'm not sure how to do this for
all
> >>elements at once (or, at least, simply). Any help is appreciated.
> >>
> >>Thanks,
> >>
> >>Tom
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
>

Hi Tom,

There is the "R Reference Card" by Tom Short and Rpad, which is 
extremely useful for a quick browsing of basic R functions. Check:

http://tolstoy.newcastle.edu.au/R/help/04/12/9637.html


For instance, colSums() and friends are in the "Matrices" section.

Best,
Dimitris

----
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/336899
Fax: +32/16/337015
Web: http://www.med.kuleuven.ac.be/biostat
     http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm


----- Original Message ----- 
From: "Mulholland, Tom" <Tom.Mulholland at dpi.wa.gov.au>
To: <r-help at stat.math.ethz.ch>
Sent: Tuesday, January 11, 2005 9:51 AM
Subject: RE: Please use colMeans()! was: Re: [R] Calculate Mean of 
ColumnVectors?

> There are indeed speed advantages in using colSums etc. However the 
> disadvantage is that the newbie doesn't always find the power 
> inherent in the apply, sapply, tapply and mapply. For many things 
> that I do, the speed is the least of my worries; although I take the 
> point that using apply for means or sums in packages that are 
> distibuted to others is not the way to go.
>
> As many of us have found out (and I think it was in S Poetry) the 
> statement was made that vectorisation befuddles some beginners. So 
> learning how to use this command, on a nice easy topic such as 
> summing or averaging does have some merit. I have to admit using 
> colSums a lot, but I don't think I have ever thought to use 
> colMeans.
>
> As an aside; does anyone have a list of optimised functions. That is 
> functions like this one whose main benefit is speed (I think) I 
> guess many of us feel that we are constantly using hammers to crack 
> the nuts, but we still don't really know what's in the toolbox.
>
> Tom
>
>> -----Original Message-----
>> From: Uwe Ligges [mailto:ligges at statistik.uni-dortmund.de]
>> Sent: Tuesday, 11 January 2005 3:57 PM
>> To: r-help at stat.math.ethz.ch
>> Cc: 'Thomas Hopper'
>> Subject: Please use colMeans()! was: Re: [R] Calculate Mean of 
>> Column
>> Vectors?
>>
>>
>> Folks, please see ?colMeans (as already pointed out in at least one
>> message) and friends.
>> These are faster than those apply(X, 2, mean) calls
>> and please do use them when publishing code (both in packages
>> in on this
>> list).
>>
>> Uwe Ligges
>>
>>
>> >>Hello,
>> >>
>> >>I've got an array defined as y <- rnorm(3000), dim(y)
<- c(3,
>> >>1000).
>> >>
>> >>I'd like to produce a 1000-element vector z that is the
mean of
>> >>the
>> >>corresponding elements of y (like z[1,1] <- mean(y[1,1],
y[2,1],
>> >>y[3,1])), but being new to R, I'm not sure how to do this
for all
>> >>elements at once (or, at least, simply). Any help is
appreciated.
>> >>
>> >>Thanks,
>> >>
>> >>Tom
>>
>> ______________________________________________
>> R-help at stat.math.ethz.ch mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide!
>> http://www.R-project.org/posting-guide.html
>>
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
>