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 at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: probable error, "weights invalid" using stset


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: probable error, "weights invalid" using stset
Date   Tue, 1 Nov 2011 12:36:37 -0400

Elizabeth <ebrown@prgs.edu> :
You are perhaps thinking of a cross-sectional regression later in the
survey, where you might want to adjust for attrition.  But if you are
using obs from 1968 on, you do not want to use the later weights...
you want to use the weights that are the inverse probability of
selection into the sample in the first place.  But things get very
complicated in the PSID, so if you want to be careful, you will have
to make your own weights to adjust for your particular sample
selection rules to make your sample representative of some larger
population--but first define that population, as I said before.  All
person-years in the US 1968 to the present?  Not possible.  All those
who were eligible or had an ancestor eligible in 1968? Etc.

On Tue, Nov 1, 2011 at 10:16 AM, Brown, Elizabeth <ebrown@prgs.edu> wrote:
> Hi Austin,  and thank you for your reply. Just a few follow-on questions; 1) Why do you recommend using first-observed and not last-observed weight? If weights adjust for attrition, wouldn't I want to use the last-observed weight for an individual, just prior to their leaving the sample? The last observed weight seems more likely to account for individual probability of leaving the sample, no? ; 2) As a clarification, you're suggesting that weights would be identified as probable error if they were 0 or missing - are there any other reasons you can think of?; 3) Last, thanks for the link to the Jenkins primer on survival analysis -- I've been traipsing all over these pages. Very helpful, thanks!
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Austin Nichols
> Sent: Monday, October 31, 2011 3:21 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: probable error, "weights invalid" using stset
>
> Elizabeth <ebrown@prgs.edu>:
> Use first-observed weights, not last, and make sure none are zero or missing. Or spend some more effort to define exactly what population you want your analysis sample to be representative of. Also note 2009 PSID data is available.  Also, you have discrete time data, not continuous time data--people are poor or not in one year, not at each point in time, at least using US definitions.  So use a discrete time hazard model; see e.g.
> http://www.iser.essex.ac.uk/resources/survival-analysis-with-stata-module-ec968
>
> On Mon, Oct 31, 2011 at 2:51 PM, Brown, Elizabeth <ebrown@prgs.edu> wrote:
>> Dear all,
>> I encountered a PROBABLE ERROR using stset that I'm unsure how to investigate/resolve. After stset, I get the following output, with a probable error with respect to 219 "weights invalid". What would cause weights to be invalid? How can I fix this?
>>
>> I'm using data from the Panel Study of Income Dynamics to examine multiple poverty transitions of families over the 1968 - 2007 period.
>>
>> With respect to how the weights were prepared, I first: 1) replaced any missing family weight values with family_weight[_n-1], and 2) made weights consistent within id by replacing  family weight with the last observed family weight value. In previous attempts to stset the data using weights, stata complained when weights were inconsistent within id.
>
> *
> *   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/
>
> __________________________________________________________________________
>
> This email message is for the sole use of the intended recipient(s) and
> may contain confidential information. Any unauthorized review, use,
> disclosure or distribution is prohibited. If you are not the intended
> recipient, please contact the sender by reply email and destroy all copies
> of the original message.
>
>
> *
> *   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