Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: logistic regression


From   "Sebastian F. B�chte" <[email protected]>
To   [email protected]
Subject   Re: st: logistic regression
Date   Mon, 12 Mar 2007 11:59:39 +0100

Looking one more time at your post I realize that you are most likely
dealing with gene data and SNP stands for "Single Nucleotide
Polymorphism". I have to say I am not familiar with the analysis of
such data but I can somehow remember that one has to be careful when
it comes to assumptions you make in advance of your analysis,eg,
independence etc.

regards
Sebastian

On 3/12/07, Sebastian F. Büchte <[email protected]> wrote:
Meena,

so you receive two measures (outcome information) from one
measurement? Do you  consider the two measure as two independet ones?

Regards
Sebastian

On 3/12/07, meena khan <[email protected]> wrote:
> Hi,
> Each individual is only measured once but the independent SNP variable has 2
> measures for each person which is why it grouped in this way. the SNP is
> binary so there are 3 possible combinations for each person (00, 01 or 11)
>
> Thanks
> Meena
>
>
> >From: "Sebastian F. Büchte" <[email protected]>
> >Reply-To: [email protected]
> >To: [email protected]
> >Subject: Re: st: logistic regression
> >Date: Mon, 12 Mar 2007 11:25:45 +0100
> >
> >Looking at your example data, your statement "only one variable (SNP)
> >is different in each person" appears to be inconsistent with the
> >presented data (check rows for id = 2).
> >
> >I would help a lot to understand how your data is generated. Multiple
> >treatments/measurements for a number of subjects? Is this perhaps
> >panel data? ...
> >
> >regards
> >sebastian
> >
> >On 3/12/07, meena khan <[email protected]> wrote:
> >>Hello,
> >>
> >>I have a dataset where all variables are replicated twice for each
> >>subject,
> >>only one variable (SNP) is different in each person:
> >>
> >>ID          SNP     Event    X1         X2
> >>
> >>1             0          0          a          a
> >>1             1          0          a          a
> >>2             1          1          b          b
> >>2             1          1          b          b
> >>
> >>I would like to know if it is correct to cluster by ID in a normal
> >>logistic
> >>regression model? Conditional logistic regression does not work as when I
> >>group it by ID as there is no variation within the groups. The logistic
> >>cluster option seems correct to me when I read the Stata help – it adjusts
> >>the SEs for intragroup correlation which is what I want. I just want to
> >>check if it is right. Thanks
> >>
> >>Meena
> >>
> >>
> >>*
> >>*   For searches and help try:
> >>*   http://www.stata.com/support/faqs/res/findit.html
> >>*   http://www.stata.com/support/statalist/faq
> >>*   http://www.ats.ucla.edu/stat/stata/
> >>
> >
> >*
> >*   For searches and help try:
> >*   http://www.stata.com/support/faqs/res/findit.html
> >*   http://www.stata.com/support/statalist/faq
> >*   http://www.ats.ucla.edu/stat/stata/
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index