In this paper a Bayesian approach to unit root testing for panel data models is proposed based on the comparison of stationary autoregressive models with and without individual deterministic trends, with their counterpart models with a unit autoregressive root. This is done under cross-sectional dependence among the units of the panel. Simulation experiments are conducted with the aim to assess...

This note examines the stochastic properties of US term spreads with parametric and semiparametric fractional integration techniques. Since the observed data (rather than the estimated residuals from a cointegrating regression) are used for the analysis, standard methods can be applied. The results indicate that US Treasury maturity rates are I(1) in most cases, although the order of integratio...

This paper investigates the effects of temporal aggregation when the aggregation frequency is variable and possibly stochastic. The results that we report include, as a particular case, the well-known results on fixed-interval aggregation, such as when monthly data is aggregated into quarters. A variable aggregation frequency implies that the aggregated process will exhibit time-varying paramet...

Are share markets too volatile? While it is difficult to ignore share market volatility it is important to determine whether volatility is excessive. This paper replicates the Shiller (1981) test as well as applying standard time series analysis to annual Australian stock market data for the period 1883 to 1999. While Shiller’s test suggests the possibility of excess volatility, time series ana...

The two-break unit root test of Lumsdaine and Papell (1997) is examined and found to suffer from bias and spurious rejections in the presence of structural breaks under the null. A two-break minimum LM unit root test is proposed as a remedy. The two-break LM test does not suffer from bias and spurious rejections and is mostly invariant to the size, location, and misspecification of the breaks. ...

This paper introduces a Lagrange Multiplier (LM) test for testing an autoregressive structure in a binary time series model proposed by Kauppi and Saikkonen (2008). Simulation results indicate that the two versions of the proposed LM test have reasonable size and power properties when the sample size is large. A parametric bootstrap method is suggested to obtain approximately correct sizes also...

We agree that either mistaking a stochastic trend for a deterministic trend (or vice-versa) is consequential for unit root tests and for tests of nonlinear serial dependence. In addition, we comment that similar results obtain for ordinary parameter inference in simple linear models. In particular, we note that detrending stochastically trended data with a deterministic polynomial or by applyin...

Log periodogram (LP) regression is shown to be consistent and to have a mixed normal limit distribution when the memory parameter d 1⁄4 1. Gaussian errors are not required. The proof relies on a new result showing that asymptotically infinite collections of discrete Fourier transforms (dft’s) of a short memory process at the fundamental frequencies in the vicinity of the origin can be treated a...

In this paper we provide empirical findings on the significance of positive feedback trading for the return behavior in the German stock market. Relying on the ShillerSentana-Wadhwani model, we use the link between index return auto-correlation and volatility to obtain a better understanding into the return characteristics generated by traders adhering to positive feedback trading strategies. O...

We show how temporal aggregation affects the size and power of the DOLS residualbased KPSS test of the null of cointegration. Size is effectively controlled by setting the minimum number of leads equal to one – as opposed to zero – when selecting the lag/lead order of the DOLS regression, but at a cost to power in finite samples. If highfrequency data for one or more series are available, we sh...