R help - Aug 2010 - Making cuts on multivariate data

I'm a newbie to R and  I was hoping someone could answer a simple question. 
I want to read in an ASCII file with 3 columns - x,y,z.  Let's say there is
a lot of data - 100,000 entries.  I then want to histogram x values that pass
arbitrary (and complicated) cuts on y and/or z.  Likewise, I want to make a
scatterplot with x and y, with a cut on z values.  Perhaps you have to redefine
the vectors first.

I'm thinking along the lines of
hist(x) for y>0 .and. sqrt(z)<4

or

plot(x,y) for x>0 .and. y>0 .and.z**2 > 100

I cant find this simple task in a first perusal of some of the tutorials.  Any
suggestions you could give would be helpful.

Thanks in advance.
Erik Ramberg

Erik Ramberg


	[[alternative HTML version deleted]]

On Aug 29, 2010, at 10:50 PM, Erik Ramberg wrote:
> I'm a newbie to R and  I was hoping someone could answer a simple  
> question.  I want to read in an ASCII file with 3 columns - x,y,z.   
> Let's say there is a lot of data - 100,000 entries.  I then want to  
> histogram x values that pass arbitrary (and complicated) cuts on y  
> and/or z.  Likewise, I want to make a scatterplot with x and y, with  
> a cut on z values.  Perhaps you have to redefine the vectors first.
>
> I'm thinking along the lines of
> hist(x) for y>0 .and. sqrt(z)<4
If the data is in a dataframe named dta with columns x,y, and z then:

with( dta[which( dta$y > 0 & sqrt(z) < 4), ],
         hist(x)  )
>
> or
>
> plot(x,y) for x>0 .and. y>0 .and.z**2 > 100
The strategy generalizes.
>
> I cant find this simple task in a first perusal of some of the  
> tutorials.
>  Any suggestions you could give would be helpful.
Don't forget to read the Posting Guide.
>
-- 
David.