Distilling Co-Movements from Persistent Macro and Financial Series

Abstract

We provide a methodology to disentangle the long-run relation between variables from their own dynamics. Macroeconomic and aggregate financial series have a high degree of inertia. If this persistence is not properly accounted for, spurious correlations will give rise to paradoxes. Our procedure shows that the Uncovered Interest Parity (UIP) puzzle evaporates when the dynamics are properly modelled: the forward premium loses all the predictive power that it seemed to have. We also show how the stock market grows in long cycles around a trend given by GDP, in a stable relation that does not break.

Keywords: ACF-based GLS procedure, Autocorrelation Function, Long memory, Nonlinearities, Uncovered Interest Parity anomaly

JEL Classification: E37