hi Gang
I have two questions about 1dSVAR (actually I have more like 100, including "what is it doing", but I will save those for another time).
1) As I mentioned before, I am extracting the timepoints from each condition and concatenating them, then giving the program the information about where the breaks in the data are. How does this affect the selection of the order of the polynomials for trend removal? In other words, these data were extracted from 15 minute+ runs, so in 3dDeconvolve we need a very high polort, but in this case that does not seem appropriate. I know it says to do the detrending within 1dSVAR (as opposed to using 3dDetrend before 1dSVAR for instance) but in this case would it make sense to use detrended data? The selection of the polort makes a big difference in the instantaneous effects.
2) I get very different results for the LR overidentification test for different subjects. How should I interpret these--i.e., is high x-squared bad and if so how bad is "too bad"? is there some cutoff I should have in mind?
thanks!
James