Skip to content

Grouping Numbers

4 messages · David Winsemius, Jorge Ivan Velez, Jason Rupert

Jason Rupert
# 
Ugh...This should be very simple, but evidently I am not searching for the proper term. 

Given the below: 
val_size<-100000
x_vals<-rnorm(val_size)

I would like to group them according to the following
x_vals_mean_tmp[1]<-mean(x_vals[1:10])
x_vals_mean_tmp[2]<-mean(x_vals[11:20])
...
x_vals_mean_tmp[n]<-mean(x_vals[99991:100000])

Then, 
I would like to group them according to the following
x_vals_mean_tmp[1]<-mean(x_vals[1:100])
x_vals_mean_tmp[2]<-mean(x_vals[101:200])
...
x_vals_mean_tmp[m]<-mean(x_vals[99901:100000])
etc.

I'm pretty sure I can come up with a loop to do this, but wondering if there is something that will allow me to break up the x_vals vector according to a certain step size.  I looked at split and cut, but those did not appear to accomplish what is needed. 

Thanks again.
David Winsemius
# 
Look at the seq function's help page.

--  
David Winsemius
On Mar 24, 2009, at 10:52 PM, Jason Rupert wrote:

            

      
      
      
    

        
Heritage Laboratories
West Hartford, CT

  
  


  


      

    

    

      
      

        
      

    

    

      
      

        


  

        

      Jason Rupert
    

    

      
      #
    

  


  

      
Jorge, 

Thank you very much for your post. 

I tried the below with a few modifications:
# First case
N<-100000
X<-rnorm(N)
step_size<-1

# Groups
g<-rep(1:(N/step_size),each=step_size)

# The result
tmp_output<-tapply(X,g,mean)

length_tmp_output<-length(tmp_output)
tmp_x_vals<-rep(step_size,length_tmp_output)
plot(tmp_x_vals, tmp_output)

for(ii in 1:val_size)
{   
step_size<-ii

# Groups
g<-rep(1:(N/step_size),each=step_size)

# The result
tmp_output<-tapply(X,g,mean)

length_tmp_output<-length(tmp_output)
tmp_x_vals<-rep(step_size,length_tmp_output)
points(tmp_x_vals, tmp_output)
}

However, when I change the step_size to 100, I receive the following error:
"Error in tapply(X, g, mean) : arguments must have same length"

Do you have any idea why the for loop crashes?  

I figured it would run smooth since it runs fine prior to the loop. 

Thanks for any insights.

        
--- On Tue, 3/24/09, Jorge Ivan Velez <jorgeivanvelez at gmail.com> wrote: