| Date: | Thu, 11 Aug 2005 10:14:50 -0500 |
|---|
| Reply-To: | Paul Thompson <paul@WUBIOS.WUSTL.EDU> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Paul Thompson <paul@WUBIOS.WUSTL.EDU> |
|---|
| Organization: | Washington University in St. Louis |
|---|
| Subject: | Re: how to combine two confidence intervals? |
|---|
|
| In-Reply-To: | <20050810131115.64235.qmail@web34304.mail.mud.yahoo.com> |
|---|
| Content-Type: | text/plain; charset=us-ascii; format=flowed |
|---|
wei yi wrote:
> Hello, All
> I have two samples, one for doctors and one for
> hospitals. I have got the confidence intervals for the
> estimators in doctors and hospitals.
> For example, doctors total payment CI is
> (120,000,000, 140,000,000) at alpha=.10 ;
> hospitals total payment CI is
> (150,000,000, 160,000,000) at alpha=0.10;
> Then I want to get one interval for the total of
> doctors and hospitals, how could I do that? Based on
> the data, I used SRS for sampling the hospital, and I
> used stratified sampling for doctors. I think direct
> add up may not be correct, or the alpha need to be
> changed.
> Thank you for your help,
> wei
>
I don't think that you are asking a very well-framed question to try to
combine them.
What will you accomplish by the combination? What future event will the
CI make predictions about?
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
|
