|
Abstract:
|
The paper shows that structural coefficients of a linear rational-expectations model (LREM) with (observed) exogenous variables and (unobserved exogenous) disturbances can be identified (uniquely determined) in terms of reduced-form coefficients of the model by solving a linear system of equations. To obtain identification, the linear-system matrix must have full rank. The paper proves this is the case if the known part of the joint coefficient matrix of exogenous variables and disturbances in the structural equation of the model has full rank. If the known part of the matrix has less than full rank, then, identification can still be obtained by adding linear or affine restrictions that give the system matrix full rank. Just as LREMs extend linear simultaneous-equations models (LSEMs) by adding terms in conditional expectations of future endogenous variables, this result extends classical identification of the coefficients of an LSEM by independent variation of exogenous variables.
|
