Marc,
Simply define the genotype variable as a variable with two categories only
(AA and AB). Forget about BB. This should solve your problem. If you really
need to know the effect of the BB genotype, just wait till you have a sample
containing the necessary number of cases to get significant results.
Hector
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Marc Busson
Sent: 28 March 2008 10:28
To:
[hidden email]
Subject: Cox
Hi all,
Do you known if it is possible to calculate a COX regression for
censured data when the number of case is equal to 0 in a categorie of a
covariate of the model.
for example:
first covariate: sex : 112 males, 85 females
second covariate: genotype : AA = 50, AB= 147, BB = 0
This give a mistake due to non convergence.
Is there a possibility to solve this problem
Thanks
--
BUSSON Marc, PhD
Inserm u662
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
