The relationship between DSGE and VAR models - cemmap
The relationship between DSGE and VAR models - cemmap
The relationship between DSGE and VAR models - cemmap
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
which can be solved using, for example, the algorithms of Blanchard <strong>and</strong> Kahn (1980), Uhlig<br />
(1999) or Sims (2002) to obtain an equation of the form (11).<br />
<strong>The</strong> next step involves choosing the set of observables, which in the AS model are<br />
Y t D .br t ;by t ; b t /: (18)<br />
Putting together (11) <strong>and</strong> (10) one can obtain the ABCD representation (2) of the loglinearized<br />
<strong>DSGE</strong> model, which is a useful starting point for discussing the <strong>relationship</strong> <strong>between</strong><br />
<strong>DSGE</strong> <strong>and</strong> <strong>VAR</strong> <strong>models</strong>.<br />
In the case of the AS model, for example, as shown by Komunjer <strong>and</strong> Ng (2011) <strong>and</strong> Morris<br />
(2012), the minimal state representation of the model (i.e., the representation of the model with<br />
the smallest number of state variables) for parameters calibrated as<br />
D .; ; ; ; ; 1 ; 2 ; r ; z ; g ; r ; z ; g / D (19)<br />
.0:995; 0:1; 53:68; 2; 1:01; 1:5; 0:5; 0:75; 0:9; 0:95; 0:002; 0:003; 0:006/<br />
is given by:<br />
2 3<br />
bz t<br />
6 bg<br />
4 t 7<br />
5<br />
br t<br />
| {z }<br />
X t<br />
2 3<br />
br t<br />
6 by<br />
4 t 7<br />
5<br />
b t<br />
| {z }<br />
Y t<br />
D<br />
D<br />
2<br />
3 2 3 2<br />
32<br />
3<br />
0:90 0 0 bz t 1 1 0 0 " zt<br />
6 0 0:95 0 7 6 bg<br />
4<br />
5 4 t 1 7<br />
5 C 6 0 1 0 76<br />
"<br />
4<br />
54<br />
gt 7 (20)<br />
5<br />
0:55 0 0:51 br t 1 0:61 0 0:69 " rt<br />
| {z }<br />
| {z } | {z }<br />
A<br />
B<br />
" t<br />
2<br />
3 2 3 2<br />
3 2 3<br />
0:90 0 0 bz t 1 1 0 0 " zt<br />
6 0 0:95 0 7 6 bg<br />
4<br />
5 4 t 1 7<br />
5 C 6 0 1 0 7 6 "<br />
4<br />
5 4 gt 7<br />
5 :<br />
0:55 0 0:51 br t 1 0:61 0 0:69 " rt<br />
| {z }<br />
| {z }<br />
C<br />
D<br />
with error covariance matrix<br />
2<br />
6 D 6<br />
4<br />
3<br />
0:003 2 0 0<br />
0 0:006 2 0 7<br />
5 : (21)<br />
0 0 0:002 2<br />
7