R help - May 2012 - select data

Dear all,

I am sure it won't be difficult for you!!
I need to calculate the average among variables for the single units of my
dataset.
But, while doing it, I need to do not consider some values.
To better explain, think like there are two units and three variables:

      V1    V2     V3
[1]   3     -2      4
[2]  -1      4      1

and you want to calculate the average by row, without considering those
negative values:

=> mean(1row) = (3+4)/2
=> mean(2row) = (4+1)/2

Could anyone please give me the commands to do that in R?

Thank you so much

	[[alternative HTML version deleted]]

This was actually discussed about a week and a half ago with many good
solutions offered, but I think the most idiomatic would be something
like this:

apply(dataset, 1, function(x) mean(x[x>0]))

The reasons I like it:

i) It uses the apply function to do the same operation row-wise
(that's what the "1" does) to all elements of your data set --
since
this is side-effect free (as a good functional language should be) it
makes for easy parallelization if you move to "big data"
ii) It uses an anonymous function (the "function ... " bit) which are
first class objects in R and can be passed as arguments to other
functions (here apply())
iii) It uses logical subscripting to pick out the values greater than
zero -- I think the subscripting behavior is the very best bit of R

Best,

Michael

On Mon, May 14, 2012 at 12:32 PM, Andrea Sica <aerdna.sica at gmail.com>
wrote:
> Dear all,
>
> I am sure it won't be difficult for you!!
> I need to calculate the average among variables for the single units of my
> dataset.
> But, while doing it, I need to do not consider some values.
> To better explain, think like there are two units and three variables:
>
> ? ? ?V1 ? ?V2 ? ? V3
> [1] ? 3 ? ? -2 ? ? ?4
> [2] ?-1 ? ? ?4 ? ? ?1
>
> and you want to calculate the average by row, without considering those
> negative values:
>
> => mean(1row) = (3+4)/2
> => mean(2row) = (4+1)/2
>
> Could anyone please give me the commands to do that in R?
>
> Thank you so much
>
> ? ? ? ?[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

This overwrites the data so you might want to create a copy first.

example <- data.frame(V1=c(3, -1), V2=c(-2, 4), V3=c(4, 1))
tf <- ifelse(example<0, TRUE, FALSE)
example[tf] <- NA
apply(example, 1, mean, na.rm=TRUE)

----------------------------------------------
David L Carlson
Associate Professor of Anthropology
Texas A&M University
College Station, TX 77843-4352
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Andrea Sica
> Sent: Monday, May 14, 2012 11:32 AM
> To: r-help at r-project.org
> Subject: [R] select data
> 
> Dear all,
> 
> I am sure it won't be difficult for you!!
> I need to calculate the average among variables for the single units of
> my
> dataset.
> But, while doing it, I need to do not consider some values.
> To better explain, think like there are two units and three variables:
> 
>       V1    V2     V3
> [1]   3     -2      4
> [2]  -1      4      1
> 
> and you want to calculate the average by row, without considering those
> negative values:
> 
> => mean(1row) = (3+4)/2
> => mean(2row) = (4+1)/2
> 
> Could anyone please give me the commands to do that in R?
> 
> Thank you so much
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.

On May 14, 2012, at 12:32 PM, Andrea Sica wrote:
> Dear all,
>
> I am sure it won't be difficult for you!!
> I need to calculate the average among variables for the single units  
> of my
> dataset.
> But, while doing it, I need to do not consider some values.
> To better explain, think like there are two units and three variables:
>
>      V1    V2     V3
> [1]   3     -2      4
> [2]  -1      4      1
>
> and you want to calculate the average by row, without considering  
> those
> negative values:
>
> => mean(1row) = (3+4)/2
> => mean(2row) = (4+1)/2
perhaps (untested in absence of reproducble example):

apply(dfrm, 1, function(x) mean(x[x>=0]) ) # would also work for a  
matrix object.
>
> Could anyone please give me the commands to do that in R?
>
> Thank you so much
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
David Winsemius, MD
West Hartford, CT