 |
 |
Guillaume et al. if you use residuals and run a new model on them, it seems to me that the dfe should be adjusted adding the rank of the 1st model performed ? Could we do that by simply changing a value in the SPM.mat? Cyril Sent from my HTC ----- Reply message ----- From: "Guillaume Flandin" <[log in to unmask]> To: <[log in to unmask]> Subject: [SPM] residuals per subject Date: Wed, Jul 15, 2015 19:04 Dear Simon, a possibility would be to call the spm_write_residuals function manually after model estimation, as the optional second argument of this function is the index of an F-contrast to adjust the data. Note also that the output data will be whitened and low-pass filtered unless you switched these off in model specification. Best regards, Guillaume. On 13/07/15 16:09, Simon Jones wrote: > Dear Guillaume > > Is there a way to do this in fMRI at the 1st level, that is without the intercept? > > Thanks, > > Simon > >> -----Original Message----- >> From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On >> Behalf Of Guillaume Flandin >> Sent: 03 February 2015 17:14 >> To: [log in to unmask] >> Subject: Re: [SPM] residuals per subject >> >> Dear Shahrzad, >> >> SPM12 allows you to choose whether to save residual images during model >> estimation: >> SPM > Stats > Model Estimation > Write residuals > Yes/No You would indeed >> create a design matrix from the 'Multiple Regression' >> option in Factorial Design specification, without adding the intercept (if you so >> wish). >> >> Best regards, >> Guillaume. >> >> >> On 03/02/15 14:18, shahrzad kharabian wrote: >>> Dear group, I need the subject-wise residual images (gray matter >>> volume) after adjusting for age, sex and ICV, to use out of SPM for >>> another analysis. >>> I would assume that I need to make a design matrix with these >>> covariates but I do not get how can one estimate the residuals per >>> subject from the standardized Betas. >>> Any input is appreciated. >>> Shahrzad >> >> -- >> Guillaume Flandin, PhD >> Wellcome Trust Centre for Neuroimaging >> University College London >> 12 Queen Square >> London WC1N 3BG > -- Guillaume Flandin, PhD Wellcome Trust Centre for Neuroimaging University College London 12 Queen Square London WC1N 3BG