A Bootstrap Method to Test Granger-Causality in the Frequency Domain

Abstract We propose a bootstrap test for unconditional and conditional Granger-causality spectra in the frequency domain. Our aims to detect if causality at particular is systematically different from zero. In particular, we consider stochastic process derived applying independently stationary original series. At each frequency, sample against distribution of median across frequencies estimated that process. Via our procedure, infer about relationship between money stock GDP Euro Area during period 1999–2017. point out aggregate M1 had significant impact on economic output all frequencies, while opposite only low frequencies.