This paper discusses a simulation procedure where innovations from time-series processes are used in conducting simulation experiments with macroeconometric models. A particular theoretical example using the term structure of interest rates is studied here, along with actual simulation experiments using a large macroeconometric model. This analysis illustrates the advantages of simulating with innovations and the extent to which more standard simulation procedures lead to misleading results. The innovation-simulation technique can be used to provide information on the response of the economy to shocks, even when the macroeconometric model is not invariant to policy changes. Policymakers might find such information to be quite valuable.

[This paper discusses a simulation procedure where innovations from time-series processes are used in conducting simulation experiments with macroeconometric models. A particular theoretical example using the term structure of interest rates is studied here, along with actual simulation experiments using a large macroeconometric model. This analysis illustrates the advantages of simulating with innovations and the extent to which more standard simulation procedures lead to misleading results. The innovation-simulation technique can be used to provide information on the response of the economy to shocks, even when the macroeconometric model is not invariant to policy changes. Policymakers might find such information to be quite valuable.]