No abstract

The panel data unit root test suggested by Levin and Lin (LL) has been widely used in several applications, notably in papers on tests of the purchasing power parity hypothesis. This test is based on a very restrictive hypothesis which is rarely ever of interest in practice. The Im–Pesaran–Shin (IPS) test relaxes the restrictive assumption of the LL test. This paper argues that although the IPS test has been offered as a generalization of the LL test, it is best viewed as a test for summarizing the evidence from a number of independent tests of the sample hypothesis. This problem has a long statistical history going back to R. A. Fisher. This paper suggests the Fisher test as a panel data unit root test, compares it with the LL and IPS tests, and the Bonferroni bounds test which is valid for correlated tests. Overall, the evidence points to the Fisher test with bootstrap‐based critical values as the preferred choice. We also suggest the use of the Fisher test for testing stationarity as the null and also in testing for cointegration in panel data.

Time series analysis has undergone many changes in recent years with the advent of unit roots and cointegration. Maddala and Kim present a comprehensive review of these important developments and examine structural change. The volume provides an analysis of unit root tests, problems with unit root testing, estimation of cointegration systems, cointegration tests, and econometric estimation with integrated regressors. The authors also present the Bayesian approach to these problems and bootstrap methods for small-sample inference. The chapters on structural change discuss the problems of unit root tests and cointegration under structural change, outliers and robust methods, the Markov-switching model and Harvey's structural time series model. Unit Roots, Cointegration and Structural Change is a major contribution to Themes in Modern Econometrics, of interest both to specialists and graduate and upper-undergraduate students.

No abstract

The paper argues that variance components models are very useful in pooling cross section and time series data because they enable us to extract some information about the regression parameters from the between group and between time-period variation-a source that is often completely eliminated in the commonly used dummy variable techniques. The paper studies the applicability and usefulness of the maximum likelihood method and analysis of covariance techniques in the analysis of this type of model, particularly when one of the covariates used is a lagged dependent variable.