Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: bootstrap weights


From   Steven Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: bootstrap weights
Date   Mon, 12 Sep 2011 18:47:14 -0400


There's no need to create a stratum variable. 

*********************
svyset _n [pweight = you weight]
*********************

will implicitly set all the data to a single stratum. However Bryan's point is well-taken. If there were clusters in the original survey design besides households, then any standard errors or CIs produced by -svy: bootstrap-, or any other survey command, will be absolutely wrong--much too small.


Steve

On Sep 12, 2011, at 12:23 PM, Bryan Sayer wrote:

If you REALLY do not have the stratum/psu information (as opposed to someone just leaving it off your file), then you can create a stratum=1 for everyone, and every observation is a single PSU, which you can generate using the observation number.

However, if your observations are people (rather than households) I would guess that at a minimum households are the PSU.  So I would check further that perhaps the information exists, but for some reason you don't have it.

Then use -svy bootstrap

Bryan Sayer
Monday to Friday, 8:30 to 5:00
Phone: (614) 442-7369
FAX:  (614) 442-7329
BSayer@chrr.osu.edu


On 9/12/2011 12:12 PM, Stefano Verde wrote:
> Dear Stata users,
> 
> I'd like to boostrap some parameters (elasticities) which I estimate
> by means of a function evaluator program (nlsur is the estimation
> method). Also, every observation has a different weight because it
> represents a different number of households in the population. So, I
> thought I could use pweights, but this is not allowed by -bootstrap-.
> 
> I guess I can't use -svy bootstrap- either because I don't have
> information on clusters nor strata.
> 
> Do you have any suggestion? I'd greatly appreciate it.
> 
> Stefano Verde
> 
> *
> *   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/


*
*   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   |   Site index