Assessing nonlinear structures in real exchange rates using recurrence plot strategies

Abstract

Purchasing Power Parity is an important theory at the basis of a large number of economic models. However, the implication derived form the theory that real exchange rates must follow stationary processes is not conclusively supported by empirical studies. In a recent paper, Serletis and Gogas (2000) show evidence of deterministic chaos in several OECD exchange rates. As a consequence, PPP rejections would be spurious. In this work, we follow a two-stage testing procedure to test for non-linearities and chaos in real exchange rates, using a new set of techniques designed by Webber and Zbilut (1994), under the denomination of Recurrence Quantification Analysis (RQA). Our conclusions differ slightly from Serletis and Gogas (2000), but they are also supportive of chaos for some exchange rates.

Recurrence Plots of Exchange Rates of Currencies

“…As told in [15], the fulfilment of the purchasing power parity is imposing equilibrium restrictions to the evolution of nominal exchange rates, and therefore, that the real exchange rates must follow a stationary process. However, the most popular linear models usually yield the finding that the real exchange rates are non-stationary [14].…”

“…As stressed in [15], several researchers considered the possibility that the underlying processes could be nonlinear stochastic or chaotic. Since it is very important to know if deterministic structures, chaotic or not, exist in the exchange rates, methods able to investigate such situation are necessary; some of them are based on recurrence analysis.…”

“…Since it is very important to know if deterministic structures, chaotic or not, exist in the exchange rates, methods able to investigate such situation are necessary; some of them are based on recurrence analysis. In [15] then, the use of recurrence plots and analysis had been considered and discussed. The aim of authors was that of checking the possibility of chaos.…”

“…The aim of authors was that of checking the possibility of chaos. In fact, a chaotic dynamics could explain why purchasing power parity does not hold for some exchange rates when applying traditional tests [15].…”

Review of Methods for the Evaluation of Human Body Balance

“…These features make it an ideal tool for analysis of COP data with respect to the above concerns. Detailed instructions for implementing RQA are presented in Belaire-Franch, Contreras and Tordera-Lledo (2002). Five measures may be obtained by RQA: percent recurrence, percent determinism, the ratio of these quantities, entropy, and trend.…”