Dear Tiziana,
The units of the averaged quantities are the same as the original units of the A, B and C parameters eg. Hz for fMRI.
If you look in bma.mEp these are the posterior means, m, and bma.sEp are the posterior standard deviations, s.
You can then get the posterior probabilities using the cumulative normal distribution eg.
post=1-spm_Ncdf(0,m,s^2)
Best, Will.
> -----Original Message-----
> From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]
> On Behalf Of Tiziana Quarto
> Sent: 29 August 2012 13:29
> To: [log in to unmask]
> Subject: [SPM] (DCM) BMA Ep
>
> Dear William and DCM experts,
> I'm currently doing DCM analyses and I'm interested in investigating
> the parameter estimates (Ep.A, B,C) of the winning model. I read the
> manual and also previous emails in this list about BMA but it's not
> clear to me how to correctly interpret the resulted values. What is the
> unit of the numbers that I obtain from this average?
> Moreover, I would like to know if it could be possible to transform
> these Ep values, obtained after the computation of the BMA, in
> posterior probabilities (indeed, I noticed that this is possible with
> the option "contrast estimates" into DCM review of the single subject).
> Your help would be very much appreciated!
> Thanks in advance,
> Tiziana
|