Skip to content

Reading Numeric Data -- Trivial Question

4 messages · Alex K, Uwe Ligges

Original
Alex K
# 
Hello,

   I am very new to R, and this is certainly and uber-newby question:

   I am trying to read a vector of numeric data that contains the log
of daily DJI returns and simply plot a histogram of it.

   The data I have is in a text file format, on each line a number
represents the log of the returns of one day.

   I have problems in reading this in a vector numeric format.

If I try

retr<- as.numeric(read.table("logDJI.TXT", header=FALSE, sep="", dec="." )); 

I get:
" Error in as.double.default(read.table("logDJIm.TXT", header = FALSE,
sep = "",  :
        (list) object cannot be coerced to 'double'"

and when I try to plot:

 plot(density(retr, width=.004), type="l", xlab="y", ylab="pdf");

I get:

"Error in density(retr, width = 0.004) : argument 'x' must be numeric"

  If I try:

retr<- as.data.frame(read.table("logDJI.TXT", header=FALSE, sep="", dec="." )); 

I get no reading or conversion error, but I get the same error when I
try to plot as above.

Can anyone help with this?

Thank you in advance,
Alex
Uwe Ligges
#
Alex K wrote:

            

      
      
      
    

        
read.table() returns a data.frame, as the docs point out, but as.numeric 
expects a vector.

Either extract the first (and only?) column or read the data using 
scan(), if there is really only one column.

Uwe Ligges

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Alex K
    

    

      
      #
    

  


  

      
On 5/20/05, Uwe Ligges <ligges at statistik.uni-dortmund.de> wrote:

            

      
      
      
    

        
Hi Uwe, thank you for the reply.

  How do I extract only the colum?

The text file looks like:
"-0.000948343416257784
0.00181733586025109
0
0.0108130251220595"

except is has about 2200 entries (numbers) instead of just four as above.

Thank you again,
Alex

            

      
      
      
    

        

      
      
    

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Uwe Ligges
    

    

      
      #
    

  


  

      
Alex K wrote:

        

            

      
      
      
    

        
Please, DO read the posting guide (cited at the bottom of each message)!
There is a manual "An Introduction to R". Please read it!
There is a manual on Data Import and Export. Please read it!

Really, why don't you read up on the hints you got in my former message? 
Why do you waste bandwith and our time with an unnecessary followup?

As I said, extract the first column of your data.frame. Hint: there are 
index operators such as in MyDataFrame[,1] or MyDataFrame[[1]].
Or use scan(). What's so difficult to read ?scan and try to apply it?

Uwe Ligges