Author(s)

Juan D. Montoro & Jose V. Paz

Abstract

As an alternative to the metric approach, two graphical tests (close returns and histogram) are implemented in Mathematica for detecting and visual- ising chaotic behaviour in macroeconomic time series. We show its ability to distinguish chaos over different simulated dynamic systems including a multisector kaldorian economic model. Finally, using them against some monetary aggregates we found no evidence of chaos in the Divisia M2 se- ries, in contrast with the results obtained in several previous works. 1 Introduction The aggregated economic activity (observed through macroeconomic time series) shows aperiodic fluctuations over time known as business cycles. Deterministic nonlinear dynamic systems allow us to generate endogenous business cycles, with some of them being able to exhibit a kind of behaviour that resembles a random process: chaotic motion

Keywords