my @a=();
my @b=();

my @a1=();
my @b1=();
[download]

Whenever you feel the need to declare variables like @a,@b,@c,... or @a1,@a2,... it is almost always better to use a container, specially if you want to extend those items without having to declare @d,@e or @a12,@a13 later on.

This also greatly simplifies your code. You want an array of columns, without having to know how many columns there will be? Use a two-dimensional array, i.e. an array of arrays. This is done with references in perl - see perlref.

open (TXT, "<$filename") || die "Can't open $filename: $!\n";
my @array;

while (my $line=<TXT>) {
    chomp($line);
    my @list = split /,/, $line;
    # now distribute the elements into each column's array
    for ( 0 .. $#list ) {
        push @{$array[$_]}, $list[$_];
    }
}
[download]

Your @array now looks like this:

@array = (
  [
    'secid',
    '002826',
    '0028262',
    '0028262',
    '0028262',
    '0028262',
    '0028262'
  ],
  [
    'Initial_shares',
    '3777',
    '3777',
    '3777',
    '3777',
    '3777',
    '3777'
  ]
);
[download]

Now populate @perl_array with the stringified elements of @array:

my @perl_array;

foreach $array_ref ( @array ) {
    my $colname = shift @$array_ref;
    my $items   = join ",", @$array_ref;
    push @perl_array, "var $colname=[$items]";
}

for ( @perl_array ) {
    print $_,"\n";
}
[download]

Output:

var secid=[002826,0028262,0028262,0028262,0028262,0028262]
var Initial_shares=[3777,3777,3777,3777,3777,3777]
[download]

Adding more columns to your data results in more elements in @perl_array. Using as input

secid,Initial_shares,Char
002826,3777,a
0028262,3777,b
0028262,3777,c
0028262,3777,d
0028262,3777,e
0028262,3777,f
[download]

results in

var secid=[002826,0028262,0028262,0028262,0028262,0028262]
var Initial_shares=[3777,3777,3777,3777,3777,3777]
var Char=[a,b,c,d,e,f]
[download]

And yes, you should use Text::CSV or such as mentioned elsewhere.