Dear Ricardo,
spm_nlsi_N and ..GN are functions that are aimed at inverting models
i.e. fitting parameters given data. As part of that process they also
generate predicted data using the current parameter values and compare
predicted to actual data. So if I understand you correctly you want to
generate predicted data given some model. This is the kind of thing
that the 'review priors' button in DCM can do (I'm not sure it
supports all the latest DCM variants but seems to work for the ERP
model). The exact code will differ depending on the kind of DCM you
are talking about but a good place to look for it would be in the
corresponding spm_dcm_{erp, ssr, csd} function below the call to
spm_nlsi_{N,GN}. You should find the line where DCM.H (or DCM.Hc) is
set (predicted responses) and then look for where the value for that
comes from. In the attached scrip you can find an example for how to
generate predicted data from an inverted DCM in the ERP case.
Best,
Vladimir
On Sun, Oct 7, 2012 at 12:10 AM, Ricardo Pizarro <[log in to unmask]> wrote:
> Hello,
>
> I am interested in using the spm_nlsi_GN.m file to generate a signal
> in the forward direction. That is by choosing the LFP or ERP model
> that specifies the differential equations, I should be able to also
> specify an input, the intrinsic and extrinsic parameters and save the
> output somewhere. My question is where in the file would I be able to
> do that? From what I understand, this file is called in order to
> specify there parameters in order to approximate a recorded signal. I
> would like to simply plot the generated output. Please email me if I
> am not explaining something correctly and thanks in advance for your
> help,
>
> --
> Ricardo Pizarro
> ---------------------------------
> Graduate Research Assistant, Meyerand Lab
> Department of Biomedical Engineering
> Wisconsin Institutes for Medical Research
> University of Wisconsin-Madison
> 1111 Highland Avenue
> Madison, WI 53705-2275
> ---------------------------------
> Cel: (301)602-0692
> Email: [log in to unmask]
|