Departures from an economic equilibrium should be mean reverting. The deviations are often assumed to be integrated of order zero but this is too restrictive. It is sufficient that the shocks are integrated of an order less than one, i.e. they may be fractionally integrated. A fractionally...

The LM type linearity test for STAR nonlinearities is severely distorted when the process is governed by conditional heteroskedasticity. In order to correct the test we propose a parametric bootstrap. It is shown, by means of Monte Carlo methods, that the bootstrap test is almost exact.

A dynamic time series LInear Structural RELation (LISREL) model is proposed for the analysis of stationary multivariate time series. The model is suitable not only for macro models, but also for panel data models. The implied covariance matrix is derived and it may be used in exact maximum...

Forecasts from seasonal cointegration models are compared with those from a standard cointegration model based on first differences and seasonal dummies. The effects of restricting or not restricting seasonal intercepts in the seasonal cointegration models are examined as well as the recently...

Uncertainty concerning future income lowers consumption. This is often called the precautionary demand for savings. In this paper the existence of precautionary saving is investigated using Swedish data for the years 1973-1992. As there are no variables for consumers' uncertainty we use proxies....

In this paper we show the consequences of applying a panel unit root test that assumes independence between the cross-sections when testing for a purchasing power parity relationship. The distribution of the tests investigated, including the IPS test of Im et al (2003), are influenced by a...

In this paper a seasonal version of the KPSS test for unit roots are proposed and its asymptotic distribution is stated. Further, a small Monte Carlo simulation is used to analyse some size and power properties.

The behaviour of Swedish stock returns over short and long run horizons is analysed. Using monthly data from 1919 to 1995 and, weekly and daily data for the 1980s and first part of the 1990s we hardly found any evidence of long run depend-ence. Using three different tests that are robust to...