 |
 |
Thanku for ur prompt reply. I will try it with lesser EV's.
Hi,
If you only have 15 subjects then you cannot fit this many EVs to this data.You need to have more data points than variables for the GLM (or any maths) to work.
All the best,Mark
On 30 Apr 2014, at 10:30, kavita singh <[log in to unmask]>wrote:
Hello,I have two groups, one patients and other control group (15 (9+6)). There are 22 beahvioral measures (variables) with 3 regressors as age,sex,education. so the total number of EV's are 22+3+2=27. (2 are control and patient).
EV1 for all controls is 1 and patients is 0. EV2 for all patients is 1 and controls is 0. rest of the EV's have demeaned values of controls+patients.
also we have 22*2=44 + 3 + 2 contrasts.
the contrasts in the glm is as such...
con ptn age sex edu var1 var2 var3 ................................var22+1 -1-1 +1+1+1+1+1-1+1-1+1-1...................................................................+1-1
this i am doing using highlevel/non-time series.As I try to view design or save GLM is shows an error of Newmat: program exception.
On Fri, Apr 25, 2014 at 11:21 PM, Mark Jenkinson <[log in to unmask]> wrote:
Hi,
You haven't given us any details on what the error message said, so I'm afraid I can't help without more information.It would also be helpful to know a little more about the input data and the design matrix and contrasts.
All the best,Mark
On 25 Apr 2014, at 15:53, kavita singh <[log in to unmask]> wrote:
Hello All,I am getting an error while am performing GLM (non-time series) with 25 EVS. I have done it but with no error in the past. Kindly help why this error arises?
Plz answer urgently.
--
Regards
Kavita Singh
Ph.D Scholar
NMR Research Centre
INMAS-DRDO
Delhi-54.
--
Regards
Kavita Singh
Ph.D Scholar
NMR Research Centre
INMAS-DRDO
Delhi-54.