Threshold Cointegration and the PPP Hypothesis

Abstract

Self-Exciting Threshold Autoregressive (SETAR) models are a non-linear variant ofconventional linear Autoregressive (AR) models. One advantage of SETAR modelsover conventional AR models lies in its flexible nature in dealing with possibleasymmetric behaviour of economic variables. The concept of threshold cointegrationimplies that the Error Correction Mechanism (ECM) at a particular interval isinactive as a result of adjustment costs, and active when deviations from equilibriumexceed certain thresholds. For instance, the presence of adjustment costs can, inmany circumstances, justify the fact that economic agents intervene to recalibrateback to a tolerable limit, as in the case when the benefits of adjustment are superiorto its costs. We introduce an approach that accounts for potential asymmetry andwe investigate the presence of the relative version of the purchasing power parity(PPP) hypothesis for 14 countries. Based on a threshold cointegration adaptation ofthe unit root test procedure suggested by Caner & Hansen (2001), we findevidence of an asymmetric adjustment for the relative version of PPP for eight pairsof countries.