Table I.
The AR and VAR coefficients across subjects (lag j = 1)
| Subject | SPM2 (AR) | VAR | |||
|---|---|---|---|---|---|
| A2L | A2R | ||||
| A 1 | A 1 | B 1 | A 1 | B 1 | |
| 1 | 0.37 | 0.30 | 0.09 | 0.08 | 0.30 |
| 2 | 0.20 | 0.31 | −0.20 | −0.03 | 0.12 |
| 3a | 0.37 | 0.78 | −0.29 | 0.74 | −0.37 |
| 4 | 0.26 | 0.42 | 0.17 | 0.02 | 0.62 |
| 5 | 0.37 | 0.61 | 0.02 | 0.20 | 0.45 |
| 6 | 0.37 | 0.39 | 0.30 | 0.17 | 0.24 |
| 7 | 0.37 | 0.22 | 0.13 | −0.02 | 0.26 |
a
Subject no. 3 needed an order 4 model to provide white residuals both for spm and standard preproccesed time courses. The A 1 coefficients when using an AR(4) model for the spm time courses of this subject were 0.5198 for A2L and 0.5173 for A2R. The VAR coefficients included in the table for subject no. 3 were estimated using an VAR(4) model.