<div dir="ltr"><div>Hello, <br><br></div><div>My name is John and I am a graduate student currently using a multivariate auto-regressive (MVAR) approach to analyze in vivo electrophysiological data which I recorded simultaneously from four subregions of a larger brain system. The goal of this set of analyses is to ask which of the subregions might be Granger-causing oscillations in specific frequency bands in the other subregions. <br><br></div><div>In short, my question is: With fieldtrip specifically, and the MVAR approach in general, is it possible to declare the dependencies in my data before constructing the MVAR model, and have that information constrain the coefficient calculation? <br><br></div><div>To elaborate, imagine the four brain regions I recorded from (A, B, C, & D) are connected unidirectionally, such that A projects to B projects to C projects to D (A->B->C->D). Using MVAR as I have been, my model will find coefficients for each of the four variables regressed on each of the other four variables. Is it possible for me to declare the dependencies before hand, such that when I build a model for B, I exclude C and D, only calculating coefficients for A and B, but when I build a model for D, I calculate coefficients for A, B, C, & D? Perhaps I have no choice but to calculate the models separately, but moreover, is it possible to declare that the brain regions are in series (unidirectionally projecting)?<br><br></div><div>Thank you for everyone's time. I apologize if I violated any rules of the community forum or if this is a variant of a previous question. I couldn't find any question similar to this one, and this question is my first to the forum. <br></div><div><br></div><div>Much obliged,<br></div><div><div><div><div class="gmail_signature"><div dir="ltr"><font style="font-family:georgia,serif" size="2"><font face="arial,helvetica,sans-serif">John</font><br></font><div><div><font style="font-family:georgia,serif" size="2"><br>__________________________________________<br></font>
<font size="2"><br>This e-mail message (including any attachments) is for the sole use of<br>
the intended recipient(s) and may contain confidential and privileged<br>
information. If the reader of this message is not the intended<br>
recipient, you are hereby notified that any dissemination, distribution<br>
or copying of this message (including any attachments) is strictly<br>
prohibited.<br><br>
If you have received this message in error, please contact
<br>
the sender by reply e-mail message and destroy all copies of the<br>
original message (including attachments).</font></div></div></div></div></div>
</div></div></div>