Statalist


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

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


From   Man Jia <jia.m@neu.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Is it normal that -svytab- works slowly?
Date   Tue, 7 Oct 2008 11:32:25 -0400 (EDT)

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/



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