Abstract:
Nonlinear models of deviations from PPP have recently provided an important, theoretically well motivated, contribution to the PPP puzzle. Most of these studies use temporally aggregated data to empirically estimate the nonlinear models. As noted by Taylor (2001), if the true DGP is nonlinear, the temporally aggregated data could exhibit misleading properties regarding the adjustment speeds. We examine the effects of different levels of temporal aggregation on\ estimates of ESTAR models of real exchange rates. Our Monte Carlo results show that temporal aggregation does not imply the disappearance of nonlinearity and that adjustment speeds are significantly slower in temporally aggregated data than in the true DGP. Furthermore, the autoregressive structure of some monthly ESTAR estimates found in the literature is suggestive that adjustment speeds are even faster than implied by the monthly estimates.