Identification of the short-run structure The identified cointegration relations are kept fixed at their previously estimated values An identified structure.

Post on 21-Dec-2015

214 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

Transcript

Identification of the short-run structure

• The identified cointegration relations are kept fixedat their previously estimated values • An identified structure should satisfy generic, empirical, and economic identification • Similar to long-run identification with one exception:• The residual covariance matrix plays an important role

How is all this connected to the ‘traditional’ idea ofidentifying a simultaneous equation system?

Formulating identifying restrictions

Interpreting shocks

Which questions?

The ‘current effects’ VAR model:

• Role of ceteris paribus• one or several competing theory models• the empirical richness of the VAR contra its theoretical ‘looseness’• the empirical irrelevance of many theory models

Example: Dornbush overshooting model

Questions to be asked in our money-demand example

Interpreting the VAR

Why are residuals correlated?

1. Causal chain effects get blurred when data are temporally aggregated

2. Expectations are inadequately modelled by the VAR model3. Omitted variable effects

A ’structural’ model

Four identification schemes

Restrictions on the reduced form model

The triangular form VAR

• Assumes a causal ordering of the variables

• The choice of ordering is generally subjective

• Sensitivity of the empirical results to the choice

Imposing general restrictions on the simultaneous effects

• Generic, empirical and economic identification

• Can we always achieve generic identification?• generic and empirical identification?• What about instrumental variables?

The rank indices of structure S1 and S2

The estimated results show that:

A partial system:

Economic identification

• Is money stock adjusting to money demand or supply?

• Is money causing prices or Are prices causing money?

• Given the empirically stable money demand relation found in the Danish data, can inflation be effectively controlled by the central bank?

• Does excess money, defined here as the deviation from the long-run money demand relation cause inflation?

top related