If you do not have any cases in one group there is no way to compare
that group to any other.
Basically the only solution to your problem is to collect more data.
Or face that your data cannot answer the problem you want to address.
Bendix Carstensne
----------------------
Bendix Carstensen
Senior Statistician
Steno Diabetes Center
Niels Steensens Vej 2
DK-2820 Gentofte
Denmark
tel: +45 44 43 87 38
mob: +45 30 75 87 38
fax: +45 44 43 07 06
[log in to unmask]
www.biostat.ku.dk/~bxc
----------------------
> -----Original Message-----
> From: A UK-based worldwide e-mail broadcast system mailing
> list [mailto:[log in to unmask]] On Behalf Of MacNeill,
> Stephanie J
> Sent: Friday, July 01, 2005 6:16 PM
> To: [log in to unmask]
> Subject: QUERY: Conditional logistic regression
>
>
> I am posting this on behalf of a colleague. All replies can
> be sent to me at [log in to unmask]
>
> Stephanie MacNeill
>
>
>
> Dear All,
>
> Using a conditional logistic regression model (stata7) I am
> analysing a case-control data set with 10 exposure groups. I
> want to generate odds ratios of each exposure group compared
> to group 10 as it has the lowest exposure). The problem is
> that group 10 does not have any cases, and therefore I can't
> generate any meaningful odds ratios. Does anyone have any
> suggestions how to solve this problem? I have considered
> adding 0.5 to each group (which is very difficult to do,
> maybe impossible, in Stata, and also may not be statistically
> sound), and combining group 10 with another group although I
> would like to avoid this.
>
>
>
|