This paper considers the properties of systems likelihood procedures for cointegrated systems when the I(2) variables are present. Two alternative methods are proposed: one is based on the full system likelihood, whereas another is based on the subsystem likelihood. By eliminating all unit roots in the system by the use of prior information concerning the presence of unit roots, these procedures yield estimates whose asymptotic distributions are mixed normal, free from nuisance parameters, and median-unbiased. Both methods are extensions of a full system maximum likelihood procedure by Phillips (1991a) to I(2) models. Three cases of cointegration with I(2) variables are considered in order to cover a wide variety of cointegration relationships. A triangular ECM representation and the two ML estimates are derived for each case, and the asymptotics are discussed as well. The asymptotic efficiency concerning the two estimates are considered.

This paper discusses inference for I(2) variables in a VAR model. The estimation procedure suggested consists of two reduced rank regressions. The asymptotic distribution of the proposed estimators of the cointegrating coefficients is mixed Gaussian, which implies that asymptotic inference can be conducted using the χ2 distribution. It is shown to what extent inference on the cointegration ranks can be conducted using the tables already prepared for the analysis of cointegration of I(1) variables. New tables are needed for the test statistics to control the size of the tests. This paper contains a multivariate test for the existence of I(2) variables. This test is illustrated using a data set consisting of U.K. and foreign prices and interest rates as well as the exchange rate.

In this article it is shown how the cointegration or joint trending behavior of economic time series helps to alleviate the errors in variables problem.

It was recently shown (Abadir, 1993b) that nonstationarity causes the limiting distributions of the Wald (W) and Lagrange multiplier (LM) statistics to become different from each other. This paper demonstrates that such a divergence between the two distributions can be used as an indicator of the presence of a unit root. A test based on this idea is devised by modifying the normalized autocorrelation coefficient (NAC). It is then shown to be an improvement on NAC in large samples and an improvement on other existing tests in large effective samples. The paper also investigates the effect of nonstationarity on the well-known inequality W ≥ LR ≥ LM.

The asymptotic variance matrix of the quantile regression estimator depends on the density of the error. For both deterministic and random regressors, the bootstrap distribution is shown to converge weakly to the limit distribution of the quantile regression estimator in probability. Thus, the confidence intervals constructed by the bootstrap percentile method have asymptotically correct coverage probabilities.

This paper presents theoretical results on the formulation and estimation of multivariate generalized ARCH models within simultaneous equations systems. A new parameterization of the multivariate ARCH process is proposed, and equivalence relations are discussed for the various ARCH parameterizations. Constraints sufficient to guarantee the positive definiteness of the conditional covariance matrices are developed, and necessary and sufficient conditions for covariance stationarity are presented. Identification and maximum likelihood estimation of the parameters in the simultaneous equations context are also covered.

The asymptotic behavior of the sample paths of two popular statistics that test market efficiency are investigated when markets learn to have rational expectations. Two cases are investigated, where, should markets start out at a rational expectations equilibrium, both statistics would asymptotically generate standard Brownian motions. In a first case, where agents are Bayesian and payoffs exogenous, the statistics have identical sample paths, but they are not standard Brownian motions. Whereas the finite-dimensional distributions are Gaussian, there may be a bias if agents' initial beliefs differ. A second case is considered, where payoffs are in part endogenous, yet agents consider them to be drawn from a stationary, exogenous distribution, which they attempt to learn in a frequentist way. In that case, one statistic behaves as if the economy were at a rational expectations equilibrium from the beginning on. The other statistic has sample paths with substantially non-Gaussian finite-dimensional distributions. Moreover, there is a negative bias. The behavior of the two statistics in the second case matches remarkably well the empirical results in an investigation of the prices of six foreign currency contracts over the period 1973–1990.