Conclusion

Abstract

This study emphasizes the advantages of interpreting the business survey data as resulting from the choices of agents who are assumed to face dynamic optimization problem. The empirical implementation of this structural analysis takes place in an errors-in-latent-variable model, which is well suited to the study of this type of data. This result is of interest to professional economists who look towards a more systematic utilization of survey data for infering the main factors at work during the business cycle. However, the methodological aspects and the economic contents of the present analysis need further improvements.