A model in which a set of dependent variables are explained by lagged values of the same set of variables. Zellner (see Garcia-Ferrer 1998) refers to VARs as “very awful regressions,” and claims that they have not been successful in forecasting.&nbsp; Allen and Fildes (2001) review evidence on VARs; it is weak. <o:p></o:p>
 <p></p>
 <ul>
   <li>García-Ferrer, A. (1998), “Professor Zellner: An 
interview for <I>the International Journal of Forecasting,” </I><SPAN 
style="mso-spacerun: yes">&nbsp;</SPAN><I>International Journal of 
Forecasting</I>, 14, 303-312.</li>
   <li>Allen, P. G. &amp; R. Fildes (2001), “Econometric 
forecasting,” in J. S. Armstrong (ed.), <I>Principles of Forecasting.</I> 
Norwell, MA: Kluwer Academic Press.</li>
 </ul>