Re: st: How to compare Somer'D in 2 groups, clustered data

[Steven Samuels <sjsamuels@gmail.com>]

Thanks, Roger! This was actually a survey data set, so -jackknife- or  -bootstrap:- with the -strata()- option should serve well.
I also thought of fitting a flexible polynomial to a -svy: logistic- model and computing the ROCs that way. 

Steve


On Mar 23, 2011, at 7:31 AM, Roger Newson wrote:

I think in this case you have to use the -bootstrap- or -jackknife- prefixes. This is because -somersd- does not automatically save delta-jackknife influence functions.

If i was to add that ability, then it would have to be done by creating an output dataset (or resultsset) with 1 observation per cluster and data on the influence function for each parameter. However, this would probably be a major upgrade for -somersd-.

I hope this helps.

Best wishes

Roger


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: r.newson@imperial.ac.uk
Web page: http://www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/popgenetics/reph/

Opinions expressed are those of the author, not of the institution.

On 21/03/2011 01:19, Steven Samuels wrote:
> I'm working on a survey with clustered weighted data.  I want to compare Somers' D  in two groups which have members in every cluster.  With Roger Newson's -somersd-, I've computed probability-weighted estimates for each of the two groups and output the cluster-based standard errors. With these, I can compute a standard error for the difference that assumes  that the groups are independent. My problem: how to take advantage of the within-cluster comparisons.  I've already tried -suest-, but that does not work with the jackknifed standard errors produced by -somersd-.  I'd appreciate suggestions.
> 
> Steve
> sjsamuels@gmail.com
> 
> 
> *
> *   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/