Asymmetric generalized impulse responses with an application in finance

Since the seminal work by Sims (1980), the impulse response functions are regularly applied to capture the propagation mechanism of a shock across time. This paper suggests a new approach for allowing asymmetry in the impulse response functions. This is an issue that has been neglected in the existing literature on the estimation of impulses. In the current paper it is shown how the underlying variables can be transformed into cumulative positive and negative changes in order to estimate the impulses to an asymmetric innovation. An application is provided to demonstrate how the propagation mechanism of an asymmetric impulse operates.