It's a numerical issue having to due with adding very large numbers with
very small ones. You don't get exactly the right number out, so there is
a small error. Over many operations, these errors accumulate.
doug
On 5/4/11 7:17 AM, soft.join Huang wrote:
Oh, sorry, Greve. I misunderstood your suggestion.
Now I get the same resulte in both mri_glmfit and matlab. Thanks a lot!
But there is still a question about how it happened. Why I have to do
so to get the correct result? There is nothing relevant between the
number scale of one variable and the beta values of other variables
theoretically. Is it any relevance to the data size of data structure
in mri_glmfit?
Best Regards,
Lijie Huang
On Tue, May 3, 2011 at 10:55 PM, Douglas N Greve
<gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu>> wrote:
sorry, I mean to do it for both the FSGD file and the matlab
script. I just tried it with your data and it works.
doug
soft.join Huang wrote:
Hi, doug.
Thanks for your suggestion, but it does not seem to work.
After having divided the ICV by 1e6, I reculated the betas,
and got the similar results as before in Matlab, just the
beta4 -- correspondent to ICV -- multiplied a 10^6.
And then I computed the rank of matrix X that the value is 4,
indicating that each 2 variables in design matrix do not
highly correlated.
Also I have finished an regress analysis including one class
variable and one continuous variable using mri_glmfit, got the
same result in Matlab.
A wired situation...
Is it any additional constrains in mri_glmfit when doing
regression analysis?
Lijie Huang
On Mon, May 2, 2011 at 9:09 PM, Douglas N Greve
<gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu>
<mailto:gr...@nmr.mgh.harvard.edu
<mailto:gr...@nmr.mgh.harvard.edu>>> wrote:
I think the problem may be that your data are "badly scaled"
meaning that one column is much greater than another
column. Try
dividing the ICV by 1e6 and see if you get more similar
results.
doug
soft.join Huang wrote:
Doug,
Thanks for your reply.
I recalculated the beta value and got the same results
as the
Matlab does, but be different from the mri_glmfit's. I
can not
figure out where the problem is.
Details following:
I use the command
mri_glmfit --table lh.volume.txt --fsgd
volume.gender.fsgd doss --C gender.diff.mtx --glmdir
lh.vol.glmdir
to finish regression analysis. and got the beta value (
beta1 = 1.9781e+03, beta2 = 2.0639e+03, beta3 = 110.9220,
beta4 = 5.1964e-04 ) from beta.mgh file.
Using X.mat file and the same dependent variable
Y.mat as
mri_glmfit's, I got totally different results in
Matlab( beta1
= 5.19e+02, beta2 = 4.44e+02, beta3 = 90.32, beta4 =
0.0015 ).
I don't know how it happened...
Hope for your reply.
The data I used are all in the attachment.
Best Regrads,
Lijie Huang
On Fri, Apr 29, 2011 at 9:45 PM,
<gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu>
<mailto:gr...@nmr.mgh.harvard.edu
<mailto:gr...@nmr.mgh.harvard.edu>>
<mailto:gr...@nmr.mgh.harvard.edu
<mailto:gr...@nmr.mgh.harvard.edu>
<mailto:gr...@nmr.mgh.harvard.edu
<mailto:gr...@nmr.mgh.harvard.edu>>>> wrote:
Lijie,
I don't know what the matlab regress() function does.
mri_glmfit
is just
solving the GLM equation, which you can do easily in
matlab:
beta = inv(X'*X)*X'*y;
try that and see if you get the same as mri_glmfit.
doug
> Hi all,
>
> I'm working on a group analysis using mri_glmfit.
> In analysis directory, I got a X.mat file which
containing the
matrix X (
> size 50x4 ) used to regress analysis, also the beta value
from
beta.mgh
> file.
> Strangely, I loaded the X.mat in Matlab and recalculated the
beta using
> function regress in matlab, but got a different result
from that of
> mri_glmfit.
> I don't know where the problem is, is there any difference in
algorithm
> between mri_glmfit and matlab function regress(
regress(Y,X) )?
or I made
> some mistakes during the procedure.
> The X.mat file and the dependent variable Y.mat are in the
attachment.
>
> Thanks in advance
>
> Lijie Huang
> _______________________________________________
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
<mailto:Freesurfer@nmr.mgh.harvard.edu>
<mailto:Freesurfer@nmr.mgh.harvard.edu
<mailto:Freesurfer@nmr.mgh.harvard.edu>>
<mailto:Freesurfer@nmr.mgh.harvard.edu
<mailto:Freesurfer@nmr.mgh.harvard.edu>
<mailto:Freesurfer@nmr.mgh.harvard.edu
<mailto:Freesurfer@nmr.mgh.harvard.edu>>>
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the
person to
whom it is
addressed. If you believe this e-mail was sent to you in
error and
the e-mail
contains patient information, please contact the
Partners
Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was
sent to
you in error
but does not contain patient information, please
contact the
sender and properly
dispose of the e-mail.
-- Douglas N. Greve, Ph.D.
MGH-NMR Center
gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu>
<mailto:gr...@nmr.mgh.harvard.edu
<mailto:gr...@nmr.mgh.harvard.edu>>
Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>
<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>
FileDrop:
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>
<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>
--
Douglas N. Greve, Ph.D.
MGH-NMR Center
gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu>
Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
