Re: [Help-bash] simmerize

[Assaf Gordon]

Hello,

> On Aug 20, 2016, at 21:54, Val Krem <address@hidden> wrote:
> 
> I am trying to summarize a huge data set by  ID and year.
> Here is the sample of data.

<sorry for he shameless plug, but>

GNU Datamash can do exactly that:
========
$ cat in.txt 
ID  Year
Ak1 2001
Ak1 2001
Ak1 2001
Ak1 2002
Ak1 2007
Ak1 2003
Ak1 2003
Ak1 2004
Ak2 2001

$ datamash --header-in -W crosstab 1,2 < in.txt
        2001    2002    2003    2004    2007
Ak1     3       1       2       1       1
Ak2     1       N/A     N/A     N/A     N/A
========

The parameters are:
  --header-in => skip the first header line in your file
  -W => whitespace-delimiters instead of TAB.
  crosstab => The operation to perform (cross-tabulation,.a.k.a. "pivot table")
  1,2 => The input fields to operate on.

This requires version 1.1.0, available here:
 https://www.gnu.org/software/datamash/
 http://ftp.gnu.org/gnu/datamash/

regards,
 - assaf