[log in to unmask] said:
> Lets say you have a regressor which is "peaks" at the onset times
> convolved with the HRF, i.e. the "standard" regressor (Kalinas first
> model). Then let us construct the regressor Kalina suggests for the
> second model (i.e. variable length, short "epochs" convolved with the
> HRF). Now we orthogonalise the second regressor with respect to the
> first and plug both into the same design matrix. Couldn't we now just
> test for the "necessity" of the orthogonalised second regressor with
> an F-test, and wouldn't that answer Kalinas question? I.e. "do I get a
> significantly better fit to my data if I take the variable response
> time into account"?
You don't need to do the orthogonalisation. That is unnecessary given
what the tests of the beta coefficients in any regression analysis do.
Each beta (with its standard error) provides a two sided t-test (or F
test) for the extra explanatory contribution of each regressor over and
above that explained by all the other regressors, and in effect by the
part of each regressor which is orthogonal to all the others. The
essential step is to use a model which incorporates the two rival
regressor models and then use the beta/t/F approach to look at the
reduction in explained sum of squares that would come about if one or
the other regressor were dropped.
Ian Nimmo-Smith
MRC Cognition and Brain Sciences Unit
15 Chaucer Road
Cambridge CB2 2EF
+44 (0) 1223 355294 x 710
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|