This note shows that a very simple model can generate returns that resemble most of the temporal and distributional behavior of long returns surprisingly well. The model is based on the stochastic unit root process introduced in Granger and Swanson (1997)

Evidence is provided in this article for the existence of a stochastic unit root (STUR) in a proxy for the US risk-free interest rate, in preference to a standard fixed unit root. The implications of the existence of the STUR, on estimating and testing the capital asset pricing model, are also...

A significant increase in the correlation coefficients of returns across countries during periods of high turbulence is regarded as evidence of the contagion of financial crises. However, heteroskedasticity is known to cause correlation coefficients to be biased upward. This note shows that...

By testing for the presence of multiple changes in persistence at a priori unknown dates in the real exchange rates spanning more than 100 years over different nominal exchange rate regimes, this note shows that the real exchange rates are more likely to be stationary during the fixed nominal...

By allowing for multiple changes in persistence, this note shows that the US/UK real exchange rate spanning two centuries is stationary. This result is consistent with the previous one in Lothian and Taylor (1996, 2000) and purchasing power parity is a useful approximation in the long run.

This article examines the validity of purchasing power parity by estimating long memory parameters with recently suggested exact local Whittle estimators of Shimotsu and Phillips (2005). Little evidence is found for stationarity in the real exchange rates spanning more than 100 years from 16...

This is a corrigendum. We correct the mistakes in Basci and Caner, "Are Real Exchange Rates Nonlinear or Non-stationary? Evidence from a New Threshold Unit Root Test" 2005, vol.9.4, Article 2.