3dLSS and autoregressive modelling


I was wondering if it is possible to run 3dLSS with input that already has undergone some autoregressive (AR) modelling.

The input to 3dLSS can be derived from 3dDeconvolve. My understanding is that 3dDeconvolve does not do very much in the way of AR modelling.

3dREMLfit looks like it would suffice, but there is no option to output a x1d file for 3dLSS, for example.

Is there a workaround for this or is it currently not possible?

Thanks for you help,


Unless you have multicollinearity issue, I suggest using 3dDeconvolve -stim_times_IM combined with 3dREMLfit instead of 3dLSS.