Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: messed up tabulate

From   "Nick Cox" <>
To   <>
Subject   RE: st: RE: messed up tabulate
Date   Fri, 28 Nov 2003 17:18:25 -0000

You're correct: the [iw=1] is wrong, and 
in fact unnecessary. And I naturally agree with your desire 
to do this directly in -tabulate- and fast. 

But, internally, that may be dependent 
on sorting as well, although I don't have sight 
of the code. 

bysort catag ined : gen freq = _N 
tabdisp catag ined, c(freq) format(%12.0gc) cellw(12) 

should reduce the overhead of interpretation 
caused by -table- by doing it directly. That's all. 


Edwin Leuven
> > Try something like 
> > 
> > bysort catag ined : gen freq = _N 
> > tabdisp catag ined [iw=1], c(freq) format(%12.0gc) cellw(12) 
> tabdisp in my stata (8.2) does not allow weights, but
> sorting 22 million observations and then calling -tabdisp- 
> isn't going to 
> shorten the time that i am waiting for results
> i appreciate that there might be *a* way, but for data 
> exploration things 
> should be easy, quick and clear
> now they are not clear that's why i'd like to appeal to 
> statacorp to fix this

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index