Eviews model in reduced form

For technical support, tips and tricks, suggestions, or any other information regarding the EViews model object.

Moderators: EViews Gareth, EViews Moderator

Posts: 3
Joined: Wed Dec 23, 2015 9:23 am

Eviews model in reduced form

Postby pappace » Fri Jun 17, 2016 6:31 am


I have some questions concering how to retrieve an Eviews model in reduced form and the reduced form parameters. The model I consider consists in a set of simultaneous equations and identities.
Some issues have been discussed here viewtopic.php?f=4&t=3832&p=51038&hilit=reduced+form#p51038, where I tried to contribute but I didn't receive answers . So, I apologise for cross-posting and I summarize my questions as follows:

1. - I have estimated a simultaneous equation system using three stage least squares (Eviews v.9). Thus, I have an object "system" in the Eviews workfile. The estimated parameters are what I consider as structural parameters.

2. - I append a set of identities to the estimated system of equations so to obtain an object "model". I am interested in solving the model in order to estimate the impact of exogenous shocks on some relevant endogenous variables.

3. - In order to solve the model, Eviews represents the model in reduced form. The model consists in a number of equations (equal to the number of endogenous variables), and each endogenous variable is expressed as a linear function of the exogenous variables. The corresponding parameters, reduced-form parameters or multipliers, are obtained as a non-linear combination of the structural ones. Then, an iterative procedure applies (for example, Gauss-Siedel) up to convergence.

4. - Here are my questions:
While it is possible to retrive the structural parameters from the object "system", it seems that it is not possible to get the reduced form parameters once the model has been solved. Please, could you suggest any way in order to retrieve the matrix of reduced form parameters?

I have a related question: let's assume that I am interested in evaluating how the model's performace changes according to a different set of reduced-form parameters, that I have computed using another tool or that I obtain from the applied literature. Is it possible to upload a different matrix of reduced form parameters and then to solve the model again? Otherwise, how do you suggest to perform this task?

Thanks a lot for your suggestions

Return to “Models”

Who is online

Users browsing this forum: No registered users and 4 guests