This all sounds very promosing Bob. I have two suggestions:
a) As I understand it, on the basis of a number of posts to this message board, the baseline model is fit to the time points that are not otherwise modeled (i.e. with other stimulus regressors). If that is correct, then would it be possible to have an option to either i) fit the baseline just like all the other stimuli (i.e. at the same time and based on all the data points), or ii) fit the baseline first, then fit the other stimulus regressors to the residual. I realise that option ii) can be done in 2 steps at the moment, but it would be nice to do in 1 step.
b) Would it be possible to include an option to specify the lags when using the minlag and maxlag options - i.e. not only have every lag between the minlag and maxlag (say 1,2,3,4,5,6 etc) but rather have a list of lags (say 1,4,7,11)?
Neither of these options is vital, so if they're too much bother, don't (bother, that is).
Thanks,
Tom