#!/bin/sh
#
RSET11=`randperm alphabet | sed 11q | tr -d '\n'`
for yob in `seq 1960 2012`
do
  gawk -v SET=$RSET11 '
      BEGIN { Numerator = 0; Denominator = 0}
        {  name = $1;  count = $2
           N = split(name,Aname,"")
           sum = 0;
           for(n =1;n<=N;n++){
               if(index(SET,Aname[n]))sum++
               else sum--
          Numerator += sum*count
          Denominator += count
 #         Denominator += count*length(name)
          }      
        }
       END {printf("%g\n", Numerator/Denominator)}
' 100yob"$yob".txt 
done >L5$RSET11