Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: How many observations can Stata 11 handle?

From   Austin Nichols <>
Subject   Re: st: How many observations can Stata 11 handle?
Date   Wed, 15 Feb 2012 09:54:12 -0500

I'm afraid I have to strongly disagree; I look forward to the day when
I can read all the classics in Stata, and I no longer have to cart
around books and e-readers.  There are only minor technical problems
to work out (e.g. limit of 65,536 codings per value label, displaying
multiple alphabets); see also help label_language but no mention of
Cyrillic there (still no Unicode).

input str15 v1 str15 v2 str15 v3
"Eh bien," "mon prince," "so Genoa"
"and Lucca" "are now" "no more than"
"family estates" "of the" "Bonapartes."
foreach v of var _all {
 encode `v', g(s`v')
 drop `v'
prog drop _all
prog def readrows
 version 8.2
 forv i=1/`=_N' {
 foreach v of var _all {
 di `"`: lab (`v') `=`v'[`i']'' "' _c

On Wed, Feb 15, 2012 at 9:28 AM, Stas Kolenikov <> wrote:

> This should suffice for variables stored as doubles (8 bytes x 3M
> observations x 20 variables = 480Mb, plus some 33% overhead). If you
> indeed have "War and Peace" coded as string variables, with one
> observation per paragraph, then things will certainly get more
> complicated, as other people have indicated, but that's hardly a good
> way to treat this masterpiece.
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index