A large set of 5350 trend following technica! trading rules is applied to LIFFE and CSCE cocoa futures prices, and to the Pound-Dollar exchange rate, in the period 1983:1-1997:6. We find that 72% of the trading rules generates positive profits, even when correcting for transaction and borrowing...

A large set of 5350 trend following technica! trading rules is applied to LIFFEand CSCE cocoa futures prices, and tothe Pound-Dollar exchange rate, in the period 1983:1-1997:6. We find that 72% ofthe trading rules generatespositive profits, even when correcting for transaction and borrowing...

We estimate a dynamic asset pricing model characterized by heterogeneous boundedly rational agents. The fundamental value of the risky asset is publicly available to all agents, but they have different beliefs about the persistence of deviations of stock prices from the fundamental benchmark. An...

This discussion paper led to a publication in the <I>Journal of Economic Dynamics and Control</I>. Volume 31(6), pp. 1938-1970.<P> We estimate a dynamic asset pricing model characterized by heterogeneous boundedly rational agents. The fundamental value of the risky asset is publicly available to all...</p></i>

This paper puts forward a method to estimate average economic growth, andits associated confidence bounds, which does not require a formal decision onpotential unit root properties. The method is based on the analysis of eitherdifference-stationary or trend-stationary time series models,...

The distribution of a functional of two correlated vector Brownian motions isapproximated by a Gamma distribution. This functional represents the limiting distribution for cointegration tests with stationary exogenous regressors, but also for cointegration tests based on a non-Gaussian...

We suggest improved tests for cointegration rank in the vector autoregressive (VAR) model and develop asymptotic distribution theory and local power results. The tests are (quasi-)likelihood ratio tests based on a Gaussian likelihood, but of course the asymptotic results apply more generally....

This paper considers tests for a unit root when the innovations follow a near-integrated GARCH process. We compare the asymptotic properties of the likelihoodratio statistic with that of the least-squares based Dickey-Fuller statistic. We first useasymptotics where the GARCH variance process is...

In this paper we present a new three-step approach to the estimation of Generalized Orthogonal GARCH (GO-GARCH) models, as proposed by van der Weide (2002). The approach only requires (non-linear) least-squares methods in combination with univariate GARCH estimation, and as such is...

It is generally believed that for the power of unit root tests, only the time span and not the observation frequency matters. In this paper we show that the observation frequency does matter when the high-frequency data display fat tails and volatility clustering, as is typically the case for...