sdm-help-list > multiple meta-regression help
Showing 1-7 of 7 posts
May 3, 2017 07:05 PM | enrico vitolo
multiple meta-regression help
Dear SDM project Team,
I'm trying to run a multiple meta-regression with patients' mean age and illness duration as regressor to evaluate the effect of these parameters on altered brain regions. When i checked the output files, I found some files (html and nifti) named as "MyRegression_10", "10m", "11", and "12"..I can't understand what these files are..according to SDM manual, when i run a meta-regression, i obtain 3 type of maps: "regressor_1" (differences between patients with maximum regressor values and healthy controls), "regressor_0" (differences between patients with minimum regressor values and healthy controls), and "regressor_1m0" (differences between patients with maximum regressor values and patients with minimum YBOCS). I can't find any indication about other type of files/maps, nor about multiple meta-regression.
What type of file/maps are the files mentioned above?
Thank you very much for your precious help!
Best regards,
Enrico Vitolo
I'm trying to run a multiple meta-regression with patients' mean age and illness duration as regressor to evaluate the effect of these parameters on altered brain regions. When i checked the output files, I found some files (html and nifti) named as "MyRegression_10", "10m", "11", and "12"..I can't understand what these files are..according to SDM manual, when i run a meta-regression, i obtain 3 type of maps: "regressor_1" (differences between patients with maximum regressor values and healthy controls), "regressor_0" (differences between patients with minimum regressor values and healthy controls), and "regressor_1m0" (differences between patients with maximum regressor values and patients with minimum YBOCS). I can't find any indication about other type of files/maps, nor about multiple meta-regression.
What type of file/maps are the files mentioned above?
Thank you very much for your precious help!
Best regards,
Enrico Vitolo
Jun 2, 2017 12:06 PM | Eunice Chen - Temple University
RE: multiple meta-regression help
Dear Enrico and Joaquim
Thank you Joaquim for making this program freely avaliable for use.
I think I have a similar problem to Enrico.
I want to check that I am interpreting multi metaregression results correctly.
So I have 2 regressors of interest. When I run the multi metaregression, I receive three outputs with these letters in the header.
01m00_z, 10m01_z and 10m00_z.
I want to check my understanding. Is this correct?
01m00_z = the effect of regressor 2, controlling for regressor 1
10m01_z = the effect of regressor 1, controlling for regressor 2
10m00_z = the effect of the interaction of regressor 1 and 2
If this isn't correct, I'd really appreciate the feedback.
Cheers, Eunice
Thank you Joaquim for making this program freely avaliable for use.
I think I have a similar problem to Enrico.
I want to check that I am interpreting multi metaregression results correctly.
So I have 2 regressors of interest. When I run the multi metaregression, I receive three outputs with these letters in the header.
01m00_z, 10m01_z and 10m00_z.
I want to check my understanding. Is this correct?
01m00_z = the effect of regressor 2, controlling for regressor 1
10m01_z = the effect of regressor 1, controlling for regressor 2
10m00_z = the effect of the interaction of regressor 1 and 2
If this isn't correct, I'd really appreciate the feedback.
Cheers, Eunice
Jun 2, 2017 02:06 PM | Joaquim Radua
RE: multiple meta-regression help
Dear Enrico and Eunice,
sorry for the confusing names of these maps, we hope that they will be less confusing in future versions of SDM.
When conducting a regression by two continuous variables:
- 10m00 is the effect of the first regressor
- 01m00 is the effect of the second regressor
Hope this helps,
Joaquim
sorry for the confusing names of these maps, we hope that they will be less confusing in future versions of SDM.
When conducting a regression by two continuous variables:
- 10m00 is the effect of the first regressor
- 01m00 is the effect of the second regressor
Hope this helps,
Joaquim
Jun 2, 2017 02:06 PM | Eunice Chen - Temple University
RE: multiple meta-regression help
Ooh thank you.
To check what does the 10m01 denote?
Cheers, Eunice
To check what does the 10m01 denote?
Cheers, Eunice
Jun 2, 2017 02:06 PM | Joaquim Radua
RE: multiple meta-regression help
Dear Eunice,
I wouldn't use 10m01 in the context of a multiple regression. This is more useful in the context of an ANOVA.
Cheers,
Joaquim
I wouldn't use 10m01 in the context of a multiple regression. This is more useful in the context of an ANOVA.
Cheers,
Joaquim
Jun 2, 2017 02:06 PM | Eunice Chen - Temple University
RE: multiple meta-regression help
ok thank you!
Jun 2, 2017 05:06 PM | enrico vitolo
RE: multiple meta-regression help
Dear Joaquim,
thank you for your help!
Your answer is really helpful.
Best regards,
Enrico
thank you for your help!
Your answer is really helpful.
Best regards,
Enrico