This paper contributes to the on-going empirical debate regarding the role of the RBC model and in particular of neutral and investmentspecific technology shocks in explaining aggregate fluctuations. To achieve this, we estimate the model’s posterior density using Bayesian
methods. Within this framework we first extend Ireland’s (2001, 2004) hybrid estimation approach to allow for a vector autoregressive moving average (VARMA) process to describe the movements and comovements of the model’s errors not explained by the basic RBC model. Our main findings for the model with neutral technical change are: (i) the VARMA specification of the errors significantly improves the hybrid model’s fit to the historical data relative to the VAR and AR alternatives; and (ii) despite setting the RBC model a more difficult task under the VARMA specification, neutral technology shocks are still capable of explaining a significant share of the observed variation
in output and its components over shorter- and longer-forecast horizons as well as hours at shorter horizons. When the hybrid model is extended to incorporate investment shocks, we find that: (iii) the VAR specification is preferred to the alternatives; and (iv) the model’s ability to explain fluctuations improves considerably.