Impulse response analysis is typically conducted by fitting an autoregression model to a time series and calculating the moving average coefficients implied by the estimated autoregression model. The possible shape and persistence of the impulse response function implied by a parsimonious autoregression specification are very limited. This paper proposes an alternative approach to estimating impulse response function, which is asymptotically valid yet is less sensitive to model misspecifications in small samples. The small sample advantages of the proposed impulse response estimator over the conventional approach is demonstrated by Monte Carlo studies. The large sample validity of the proposed estimator is also established.
Chang, Pao Li and Sakata, Shinichi.
A Misspecification-Robust Impulse Response Estimator. (2002). Research Collection School Of Economics.
Available at: http://ink.library.smu.edu.sg/soe_research/688
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.