PANEL INDEX VAR MODELS: SPECIFICATION, ESTIMATION - Ivie
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
B o1 | Y T ẑ 3<br />
, ψ −Bo1 ∼ IG<br />
2 , ˆQ<br />
<br />
3<br />
;<br />
2<br />
<br />
Boi<br />
−1 | Y T , ψ −Boi ∼ W<br />
ẑ 4i , ˆQ 4i ;<br />
Ψ −1 | Y T , ψ −Ψ ∼ W<br />
ẑ o , ˆQ<br />
<br />
o .<br />
where expressions for ˆδ t , ˆV t , ẑ 1 , ˆQ 1 , ˆθ o , ˆΨ −1 , ˆµ, ˆΣ µ ,ẑ 2 , ˆQ 2 , ẑ 3 , ˆQ 3 , ẑ 4i , ˆQ 4i , ẑ o , ˆQ o are given in the<br />
appendix.<br />
Depending on the application, the conditional posterior of (θ 1 , ..., θ T | Y T , ψ −θt ), can be<br />
obtained recursively either with the Kalman filter or the Kalman smoother, as in Chib and<br />
Greenberg (1995). In the first case, we initialize {θ t } t<br />
for each t and save:<br />
ˆθ t|t = ˆθ<br />
<br />
t|t−1 ∗ + K t δ t − Ξˆθ t|t−1<br />
∗ (13)<br />
R t|t = (I − K t Ξ) Rt|t−1<br />
∗<br />
K t = Rt|t−1ΞF ∗ −1<br />
t|t−1<br />
F t|t−1 = ΞRt|t−1Ξ ∗ + B 1<br />
where ˆθ t|t−1 ∗ = ˆθ t−1|t−1 ∗ and R∗ t|t−1 = R∗ t−1|t−1 +ξ tB o ,andˆθ<br />
t−1|t−1 ∗ and R∗ t−1|t−1<br />
are, respectively,<br />
the mean and the variance covariance matrix of the conditional distribution of θt−1|t−1 ∗ .<br />
Draws from for θ t aremadefromN(ˆθ t|t ,R t|t ). In the second case, the conditional posterior<br />
of θ 1 , ..., θ T | Y T , ψ −θt is sampled in reverse time order from<br />
θ T<br />
<br />
∼ N<br />
ˆθT |T ,R T |T<br />
θ T −1<br />
<br />
∼ N<br />
ˆθT −1 ,R T −1 (14)<br />
. <br />
θ 1 ∼ N<br />
ˆθ1 ,R 1<br />
where ˆθ t = ˆθ t|t + Ξ t<br />
<br />
θ t+1 − ˆθ t|t<br />
<br />
, R t = R t|t − Ξ t R ∗ t+1|t Ξ t,andΞ t = R t|t R ∗−1<br />
t+1|t .<br />
To make the updating scheme described in (13)-(14) operational, initial values at time<br />
t = 1 must be assigned. For instance, one can choose to initialize B 0 = R 0 to be diagonal with<br />
elements φ i equal to small values. ˆθ0 can be initialized by running a <strong>VAR</strong> for each country<br />
and taking the constant. In the same way, Q 1 can be taken as the variance covariance matrix<br />
of a pooled <strong>VAR</strong>, and Ω can be initialized by setting it equal to Q 1 .<br />
11