We estimate exponential smooth transition autoregressive (ESTAR) models of deviations from PPP, which are obtained using the Johansen cointegration method, for both CPI-and WPIbased measures and a broad set of U.S. trading partners. In several cases, we nd clear evidence of a mean-reverting dynamic process for sizable deviations from PPP, with the equilibrium tendency varying nonlinearly with the magnitude of disequilibrium. Analysis of impulse response functions also supports a nonlinear dynamic structure, but convergence to long-run PPP in the post-Bretton Woods era is very slow.

Tests are made of the stochastic long memory in the Greek stock market, an emerging capital market. The fractional differencing parameter is estimated using the spectral regression method. Contrary to findings for major capital markets, significant and robust evidence of positive long-term persistence is found in the Greek stock market. As compared to benchmark linear models, the estimated fractional models provide improved out-of-sample forecasting accuracy for the Greek stock returns series over longer forecasting horizons.

We test for long term dependence in U.S. stock returns, analyzing composite and sectoral stock indices and firms' returns series to evaluate aggregation effects. Fractal dynamics are not detected in stock indices, but are present in some firms' returns series.

This paper investigates the effects of exchange rate uncertainty on the volume and variability of trade flows. Employing a signal extraction framework, we show t h a t the direction and magnitude of importers' and exporters' optimal trading activities depend upon the source of the uncertainty (general microstructure shocks, fundamental factors driving the exchange rate process, or a noisy signal of policy innovations), providing a rationale for the contradictory empirical evidence in the literature. W e also show that exchange rate uncertainty emanating from general microstructure shocks and the fundamental factors reduces the variability of trade flows, while that related to a noisy signal of policy innovations increases variability.

We are grateful to Pedro de Lima and Uwe Hassler for providing their software and assistance in its modification. We acknowledge the helpful comments of two anonymous reviewers and a coeditor of this journal. The standard disclaimer applies.

No abstract

No abstract

This paper tests for fractional roots in the futures prices for selected commodities, foreign currencies, and stock indexes. The fractional testing method is the spectral regression method suggested by Geweke and Porter-Hudak (1983). The empirical results suggest the presence of a fractional exponent in the differencing process for several commodity and foreign currency futures prices. The returns series for these commodities and currencies exhibit long range positive dependence. However, differencing of exact order one is sufficient for the stock index futures prices. Implications are drawn concerning theoretical and econometric modeling and price forecasting.