Working Papers

• What Can We Gain from Pooling VAR and Local Projection Impulse Response Functions?

(Job Market Paper)

Abstract: I show that pooling Vector Autoregression (VAR) and Local Projection (LP) impulse response functions (IRFs) may address the bias-variance trade-off in IRF estimation. I also propose a ``sequence-of-block bootstrap'' method, designed to construct confidence intervals for the pooled IRFs. I document three key findings: (i) Monte Carlo exercises suggest that the pooling approach outperforms both VAR and LP in terms of root mean squared error. (ii) Using the sequence-of-block bootstrap, inference based on the pooled estimator delivers coverage rates that are slightly lower than those under LP and more accurate than those based on VAR, while yielding substantially shorter average interval lengths than LP. (iii) Finally, the pooling approach combined with the sequence-of-block bootstrap produces sensible results in an empirical application to monetary policy.