Statalist


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

st: RE: Is it normal that -svytab- works slowly?


From   "Martin Weiss" <martin.weiss1@gmx.de>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Is it normal that -svytab- works slowly?
Date   Tue, 7 Oct 2008 17:40:50 +0200

"...,although the memory has already been set to 1g." This makes me think that you might have fallen into the trap of thinking that raising the memory with -set mem- will make Stata faster which is quite wrong. -set mem- allocates more memory to your data which detracts memory for other (Stata) purposes...
Passing off your pweights as fweights will be impossible if they are not integers, by the way...


HTH
Martin


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Man Jia
Sent: Tuesday, October 07, 2008 5:32 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: Is it normal that -svytab- works slowly?

Hi all,

I was wondering if anyone could give me some suggestion on command 
-svytab-.

My goal is to get the frequencies distribution of most of the variables 
in a survey  data set (with sampling weight). I don't choose -tabulate- 
because it can't be used together with pweight.
So, I generate a constant variable which equals 1 and -svytab each 
variable I'm interested with in the data set together with the constant 
variable 1. 

The problem is that, it takes about 5 minutes to do just one svytab . 
After svytab for 4 variables, stata shows there's not enough memory to 
do svytab to the rest variables,although the memory has already been 
set to 1g. (The data set has 3,932,343 observations and 90 variables.)

So, my questions are as follows.
(1) Is it normal that -svytab- need so much memory and it works so 
slowly ? 
(2) Is there any better way to check frequencies of a lot of variables 
with sampling weight?
(3) Suppose the goal of checking frequencies is just to see  if there's 
anything weird or any errors in the data. Is there any problem if I 
just use -tabulate- and put the weight variable  as aweight or fweight 
instead of pweight? 

Thanks a lot for your help!

Mandy
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index