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

st: RE: re: Sample with weights

From   "Willard van Ooij" <>
To   <>
Subject   st: RE: re: Sample with weights
Date   Mon, 3 Oct 2005 10:15:54 +0200

Mike, Nick and Svend,

Thanks very much for your suggestions. I will look into them and see
what works best for my case.


-----Oorspronkelijk bericht-----
Van: Mike Lacy [] 
Verzonden: zaterdag 1 oktober 2005 17:05
Onderwerp: st: re: Sample with weights

>Date: Fri, 30 Sep 2005 11:48:40 +0200
>From: "Willard van Ooij" <>
>Subject: st: Sample with weights
>Dear statalisters,
>I have a population of company's. I want a sample from this population,

>but the probability of a company to be sampled has to be equivalent 
>with the number of employees (let's call this "size").
>So I thought i could
>- -sample 10 [fweight=size]
>To get a 10 percent sample. But -sample- doesn't accept weights. A 
>solution might be to expand the dataset with size, but then a company 
>can get sampled several times if it has more than one employee, and I 
>don't want that to happen. I think there must be a very simple 
>solution, but I haven't been able to find it.

This is simple, produces a sample of exactly the desired size, and I
believe fulfills the condition of the probability of selection being
proportional to size . *Assume "Size" is the company size variable, and
M is the desired sample size gen ppsorder = uniform() * Size sort
ppsorder keep if _n <= M drop ppsorder

Yes, sorting the file is a bit clumsy, but this is presumably a one time
thing, not something appearing inside a loop.


Mike Lacy
Fort Collins CO USA
(970) 491-6721 office

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index