Dear all,
does anyone know one or two good method papers to cite when describing the difference of DOSS and DODS? I have already screened Pubmed and Co. but did not really find anything convincing.
Any help greatly appreciated!
Best wishes, Christian Scheel
It really is just a matter of justifying and explaining your chosen statistical model. If you believe that two groups have the same regression slope for a parameter (but possibly different offsets/intercepts), then you would use a "DOSS" model -- this is akin to a model in SPSS/SAS in which you model a main effect of group and a single (main) effect of the regressor that applies to all subjects (e.g., Y = group age). If you believe that the two groups might possibly have a different slope, then you would use a "DODS" model (also known as a heterogeneous slopes model) -- this is equivalent to a model in SPSS/SAS in which you model a main effect of group (different offsets/intercepts) and separate (heterogeneous) slopes for the two groups i.e,. you model a possible interaction between group and slope; Y = group age group*age).
cheers, -MH
On Tue, 2011-02-01 at 20:49 +0100, Christian Scheel wrote:
Dear all,
does anyone know one or two good method papers to cite when describing the difference of DOSS and DODS? I have already screened Pubmed and Co. but did not really find anything convincing.
Any help greatly appreciated!
Best wishes, Christian Scheel
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.
Thank you Michael for your explanations,
that is exactly the way I understood DODS and DOSS. I am planning to present DOSS and DODS results in comparison and wanted to explain both models in a similar manner that you have done it here. But as these DOSS and DODS terms are kind of unique to Freesurfer and might not be known to readers not using Freesurfer I felt I had to show where I found these expressions and that I did not invent these terms.
Best wishes, Christian Scheel
Am 01.02.2011 22:18, schrieb Michael Harms:
It really is just a matter of justifying and explaining your chosen statistical model. If you believe that two groups have the same regression slope for a parameter (but possibly different offsets/intercepts), then you would use a "DOSS" model -- this is akin to a model in SPSS/SAS in which you model a main effect of group and a single (main) effect of the regressor that applies to all subjects (e.g., Y = group age). If you believe that the two groups might possibly have a different slope, then you would use a "DODS" model (also known as a heterogeneous slopes model) -- this is equivalent to a model in SPSS/SAS in which you model a main effect of group (different offsets/intercepts) and separate (heterogeneous) slopes for the two groups i.e,. you model a possible interaction between group and slope; Y = group age group*age).
cheers, -MH
On Tue, 2011-02-01 at 20:49 +0100, Christian Scheel wrote:
Dear all,
does anyone know one or two good method papers to cite when describing the difference of DOSS and DODS? I have already screened Pubmed and Co. but did not really find anything convincing.
Any help greatly appreciated!
Best wishes, Christian Scheel
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.
Hi Christian, yes, these are unique! I came up with them in about 30 seconds when I wrote mri_glmfit about 10 years ago. If I had known they were going to last this long, I would have spent more time coming up with a good name for them!
doug
Christian Scheel wrote:
Thank you Michael for your explanations,
that is exactly the way I understood DODS and DOSS. I am planning to present DOSS and DODS results in comparison and wanted to explain both models in a similar manner that you have done it here. But as these DOSS and DODS terms are kind of unique to Freesurfer and might not be known to readers not using Freesurfer I felt I had to show where I found these expressions and that I did not invent these terms.
Best wishes, Christian Scheel
Am 01.02.2011 22:18, schrieb Michael Harms:
It really is just a matter of justifying and explaining your chosen statistical model. If you believe that two groups have the same regression slope for a parameter (but possibly different offsets/intercepts), then you would use a "DOSS" model -- this is akin to a model in SPSS/SAS in which you model a main effect of group and a single (main) effect of the regressor that applies to all subjects (e.g., Y = group age). If you believe that the two groups might possibly have a different slope, then you would use a "DODS" model (also known as a heterogeneous slopes model) -- this is equivalent to a model in SPSS/SAS in which you model a main effect of group (different offsets/intercepts) and separate (heterogeneous) slopes for the two groups i.e,. you model a possible interaction between group and slope; Y = group age group*age).
cheers, -MH
On Tue, 2011-02-01 at 20:49 +0100, Christian Scheel wrote:
Dear all,
does anyone know one or two good method papers to cite when describing the difference of DOSS and DODS? I have already screened Pubmed and Co. but did not really find anything convincing.
Any help greatly appreciated!
Best wishes, Christian Scheel
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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Doug,
OK, I see. Honestly, I think the names could be much worse! Really not bad for 30 seconds... ;-) But as I understand there is no paper of you naming DOSS and DODS that I could cite, correct? I will try to find a different solution then.
Thanks for your infos!
Best wishes, Christian Scheel
---
Am 01.02.2011 23:29, schrieb Douglas N Greve:
Hi Christian, yes, these are unique! I came up with them in about 30 seconds when I wrote mri_glmfit about 10 years ago. If I had known they were going to last this long, I would have spent more time coming up with a good name for them!
doug
Christian Scheel wrote:
Thank you Michael for your explanations,
that is exactly the way I understood DODS and DOSS. I am planning to present DOSS and DODS results in comparison and wanted to explain both models in a similar manner that you have done it here. But as these DOSS and DODS terms are kind of unique to Freesurfer and might not be known to readers not using Freesurfer I felt I had to show where I found these expressions and that I did not invent these terms.
Best wishes, Christian Scheel
Am 01.02.2011 22:18, schrieb Michael Harms:
It really is just a matter of justifying and explaining your chosen statistical model. If you believe that two groups have the same regression slope for a parameter (but possibly different offsets/intercepts), then you would use a "DOSS" model -- this is akin to a model in SPSS/SAS in which you model a main effect of group and a single (main) effect of the regressor that applies to all subjects (e.g., Y = group age). If you believe that the two groups might possibly have a different slope, then you would use a "DODS" model (also known as a heterogeneous slopes model) -- this is equivalent to a model in SPSS/SAS in which you model a main effect of group (different offsets/intercepts) and separate (heterogeneous) slopes for the two groups i.e,. you model a possible interaction between group and slope; Y = group age group*age).
cheers, -MH
On Tue, 2011-02-01 at 20:49 +0100, Christian Scheel wrote:
Dear all,
does anyone know one or two good method papers to cite when describing the difference of DOSS and DODS? I have already screened Pubmed and Co. but did not really find anything convincing.
Any help greatly appreciated!
Best wishes, Christian Scheel
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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
right, there is no paper. As Michael points out, these are fairly standard stats models.
doug
Christian Scheel wrote:
Hi Doug,
OK, I see. Honestly, I think the names could be much worse! Really not bad for 30 seconds... ;-) But as I understand there is no paper of you naming DOSS and DODS that I could cite, correct? I will try to find a different solution then.
Thanks for your infos!
Best wishes, Christian Scheel
Am 01.02.2011 23:29, schrieb Douglas N Greve:
Hi Christian, yes, these are unique! I came up with them in about 30 seconds when I wrote mri_glmfit about 10 years ago. If I had known they were going to last this long, I would have spent more time coming up with a good name for them!
doug
Christian Scheel wrote:
Thank you Michael for your explanations,
that is exactly the way I understood DODS and DOSS. I am planning to present DOSS and DODS results in comparison and wanted to explain both models in a similar manner that you have done it here. But as these DOSS and DODS terms are kind of unique to Freesurfer and might not be known to readers not using Freesurfer I felt I had to show where I found these expressions and that I did not invent these terms.
Best wishes, Christian Scheel
Am 01.02.2011 22:18, schrieb Michael Harms:
It really is just a matter of justifying and explaining your chosen statistical model. If you believe that two groups have the same regression slope for a parameter (but possibly different offsets/intercepts), then you would use a "DOSS" model -- this is akin to a model in SPSS/SAS in which you model a main effect of group and a single (main) effect of the regressor that applies to all subjects (e.g., Y = group age). If you believe that the two groups might possibly have a different slope, then you would use a "DODS" model (also known as a heterogeneous slopes model) -- this is equivalent to a model in SPSS/SAS in which you model a main effect of group (different offsets/intercepts) and separate (heterogeneous) slopes for the two groups i.e,. you model a possible interaction between group and slope; Y = group age group*age).
cheers, -MH
On Tue, 2011-02-01 at 20:49 +0100, Christian Scheel wrote:
Dear all,
does anyone know one or two good method papers to cite when describing the difference of DOSS and DODS? I have already screened Pubmed and Co. but did not really find anything convincing.
Any help greatly appreciated!
Best wishes, Christian Scheel
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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu