Understanding macroeconomic models: Structural sensitivity analysis of a medium-sized model

To get a better impression of the quantitative relationships in/of the various channels of macroeconomic models, Kuh, Neese, and Hollinger introduced the technique of systematic parameter perturbation. This technique is applied to the RWI-business cycle model, a medium sized (41 stochastic equations, 86 definitions), quarterly macroeconometric model for the FRG. The evaluation of the results concentrates on (1) the sensitivity of the model to parameter perturbations in general, and to (2) the sensitivity of policy goal variables in particular. The findings show that in the model the number of important within-block and between-block relationships is much smaller than suggested by usual incidence matrices, providing additional evidence for Simon's empty world hypothesis.Paper presented at the Annual Meeting of the Society for Economic Dynamics and Control, Anacapri, Italy, 18–20 June, 1991.