The autoregressive fractionally integrated moving average (ARFIMA) model has become a popular approach for analyzing time series that exhibit long-range dependence. For the Gaussian case, there have been substantial advances in the area of likelihood-based inference, including development of the asymptotic properties of the maximum likelihood estimates and formulation of procedures for their computation. Small-sample inference, however, has not to date been studied. Here we investigate the small-sample behavior of the conventional and Bartlett-corrected likelihood ratio tests (LRT) for the fractional difference parameter. We derive an expression for the Bartlett correction factor. We investigate the asymptotic order of approximation of the Bartlett-corrected test. In addition, we present a small simulation study of the conventional and Bartlett-corrected LRT's. We find that for simple ARFIMA models both tests perform fairly well with a sample size of 40 but the Bartlett-corrected test generally provides an improvement over the conventional test with a sample size of 20.

Testing the cointegrating rank of a vector autoregressive process with an intercept is considered. In addition to the likelihood ratio (LR) tests developed by Johansen and Juselius (1990, Oxford Bulletin of Economics and Statistics, 52, 169–210) and others we also consider an alternative class of tests that is based on estimating the trend parameters of the deterministic term in a different way. The asymptotic local power of these tests is derived and compared to that of the corresponding LR tests. The small sample properties are investigated by simulations. The new tests are seen to be substantially more powerful than conventional LR tests.

This paper proposes a semiparametric estimator for multiple equations multiple index (MEMI) models. Examples of MEMI models include several sample selection models and the multinomial choice model. The proposed estimator minimizes the average distance between the dependent variable unconditional and conditional on an index. The estimator is √N-consistent and asymptotically normally distributed. The paper also provides a Monte Carlo experiment to evaluate the finite-sample performance of the estimator.

The classical definitions of GARCH-type processes rely on strong assumptions on the first two conditional moments. The common practice in empirical studies, however, has been to test for GARCH by detecting serial correlations in the squared regression errors. This can be problematic because such autocorrelation structures are compatible with severe misspecifications of the standard GARCH. Numerous examples are provided in the paper. In consequence, standard (quasi-) maximum likelihood procedures can be inconsistent if the conditional first two moments are misspecified. To alleviate these problems of possible misspecification, we consider weak GARCH representations characterized by an ARMA structure for the squared error terms. The weak GARCH representation eliminates the need for correct specification of the first two conditional moments. The parameters of the representation are estimated via two-stage least squares. The estimator is shown to be consistent and asymptotically normal. Forecasting issues are also addressed.

The paper considers different versions of the Lagrange multiplier (LM) tests for autocorrelation and/or for conditional heteroskedasticity. These versions differ in terms of the residuals, and of the functions of the residuals, used to build the tests. In particular, we compare ordinary least squares versus least absolute deviation (LAD) residuals, and we compare squared residuals versus their absolute value. We show that the LM tests based on LAD residuals are asymptotically distributed as a χ2 and that these tests are robust to nonnormality. The Monte Carlo study provides evidence in favor of the LAD residuals, and of the absolute value of the LAD residuals, to build the LM tests here discussed.

This paper develops a new estimation method for nonstationary vector autoregressions (VAR's) with unknown mixtures of I(0), I(1), and I(2) components. The method does not require prior knowledge on the exact number and location of unit roots in the system. It is, therefore, applicable for VAR's with any mixture of I(0), I(1), and I(2) variables, which may be cointegrated in any form. The limit theory for the stationary component of our estimator is still normal, thereby preserving the usual VAR limit theory. Yet, the leading term of the nonstationary component of the estimator has mixed normal limit distribution and does not involve unit root distribution. Our method is an extension of the FM-VAR procedure by Phillips (1995, Econometrica 63, 1023–1078) and yields an estimator that is optimal in the sense of Phillips (1991, Econometrica 59, 283–306). Moreover, we show for a certain class of linear restrictions that the Wald tests based on the estimator are asymptotically distributed as a weighted sum of independent chi-square variates with weights between zero and one. For such restrictions, the limit distribution of the standard Wald test is nonstandard and nuisance parameter dependent. This has a direct application for Granger-causality testing in nonstationary VAR's.

This paper investigates Granger noncausality and the cointegrating relation between two time series in the Hilbert space framework. This framework allows us to analyze the relationship between cointegration and distance between two information sets. In particular, we prove that if two variables, X and Y, are cointegrated, then the distance between two information sets, concerning the differenced series ΔX and ΔY, must be less than the standard deviation of ΔX.

In this paper I present an alternative derivation of the asymptotic distribution of Kremers, Ericsson, and Dolado's (1992, Oxford Bulletin of Economics and Statistics 54, 325–348) conditional error correction model (ECM)–based t-test for cointegration with a single prespecified cointegrating vector. This alternative distribution, which is identical to the distribution of Hansen's (1995, Econometric Theory 11, 1148–1171) covariate augmented t-test for a unit root, is valid for weakly exogenous regressors and depends on a consistently estimable nuisance parameter that takes on values in the unit interval. I show analytically, using asymptotic power functions based on near-cointegrated alternatives, that the ECM t-test with a prespecified cointegrating vector can have much higher power than single equation tests for cointegration based on estimating the cointegrating vector. I also characterize situations in which the ECM t-test computed with a misspecified cointegrating vector will have high power.

A procedure for testing the significance of a subset of explanatory variables in a nonparametric regression is proposed. Our test statistic uses the kernel method. Under the null hypothesis of no effect of the variables under test, we show that our test statistic has an nhp2/2 standard normal limiting distribution, where p2 is the dimension of the complete set of regressors. Our test is one-sided, consistent against all alternatives and detects local alternatives approaching the null at rate slower than n−1/2h−p2/4. Our Monte-Carlo experiments indicate that it outperforms the test proposed by Fan and Li (1996, Econometrica 64, 865–890).

Consider the model y = x′β0 + f*(z) + ε, where ε [d over =] N(0, σ02). We calculate the smallest asymptotic variance that n1/2 consistent regular (n1/2CR) estimators of β0 can have when the only information we possess about f* is that it has a certain shape. We focus on three particular cases: (i) when f* is homogeneous of degree r, (ii) when f* is concave, (iii) when f* is decreasing. Our results show that in the class of all n1/2CR estimators of β0, homogeneity of f* may lead to substantial asymptotic efficiency gains in estimating β0. In contrast, at least asymptotically, concavity and monotonicity of f* do not help in estimating β0 more efficiently, at least for n1/2CR estimators of β0.

Professor Olav Reiersøl was born on 28 June 1908 in the countryside in Norway. He enrolled at Oslo University in 1928, where he studied chemistry, mathematics, physics, and botany. Choosing mathematics as his main subject, he received the cand. real. degree in 1935. As a graduate in mathematics and mathematical statistics he was employed as an assistant by Professor Ragnar Frisch.

The prediction of future events and developments is an exciting and perhaps mysterious task, often associated with the aura of prophets and seers instead of probabilistic models and computer screens. The reality of macroeconomic forecasting, however, is quite mundane. Predictions of macroeconomic aggregates play an important role in the decision making of private enterprises, central banks, and governments. In general, forecasts become less popular if they turn out to be inaccurate ex post, and the postwar history of macroeconomic forecasting has had its share of disappointments. For instance, in the early 1980's, economists tested inflation forecasts taken over the previous 20 years and found that the forecasts were poor, partly as a result of the oil price shocks in the 1970's. A recent study (Croushore, 1998) with data up to 1996 provides a more favorable assessment of the quality of inflation forecasts.

Misclassification in binary choice (binomial response) models occurs when the dependent variable is measured with error, that is, when an actual “one” response is sometimes recorded as a zero and vice versa. This paper shows that binary response models with misclassification are semiparametrically identified, even when the probabilities of misclassification depend in unknown ways on model covariates and the distribution of the errors is unknown.

A strong consistency result for heteroskedasticity and autocorrelation consistent covariance matrix estimators is proven in this paper. In addition, an error in a weak consistency proof for such estimators in the econometrics literature and a correction of that result is provided.

Time series data are often well modeled by using the device of an autoregressive root that is local to unity. Unfortunately, the localizing parameter (c) is not consistently estimable using existing time series econometric techniques and the lack of a consistent estimator complicates inference. This paper develops procedures for the estimation of a common localizing parameter using panel data. Pooling information across individuals in a panel aids the identification and estimation of the localizing parameter and leads to consistent estimation in simple panel models. However, in the important case of models with concomitant deterministic trends, it is shown that pooled panel estimators of the localizing parameter are asymptotically biased. Some techniques are developed to overcome this difficulty, and consistent estimators of c in the region c < 0 are developed for panel models with deterministic and stochastic trends. A limit distribution theory is also established, and test statistics are constructed for exploring interesting hypotheses, such as the equivalence of local to unity parameters across subgroups of the population. The methods are applied to the empirically important problem of the efficient extraction of deterministic trends. They are also shown to deliver consistent estimates of distancing parameters in nonstationary panel models where the initial conditions are in the distant past. In the development of the asymptotic theory this paper makes use of both sequential and joint limit approaches. An important limitation in the operation of the joint asymptotics that is sometimes needed in our development is the rate condition n/T → 0. So the results in the paper are likely to be most relevant in panels where T is large and n is moderately large.

Likelihood ratio tests for restrictions on cointegrating vectors are asymptotically χ2 distributed. For some values of the parameters this asymptotic distribution does not give a good approximation to the finite sample distribution. In this paper we derive the Bartlett correction factor for the likelihood ratio test and show by some simulation experiments that it can be a useful tool for making inference.

This paper compares the performance of several single and system estimators of a two equation simultaneous model with unbalanced panel data. The Monte Carlo design varies the degree of unbalancedness in the data and the variance components ratio due to the individual effects. One useful result for applied researchers is that the feasible error components 2SLS and 3SLS procedures based on simple ANOVA type estimators of the variance components perform well with incomplete panels and are recommended in practice.

Benedikt Pötscher and Ingmar Prucha are two exceptional econometricians who combine an extraordinary knowledge of the statistics and econometrics literature with great analytical skills. Both are excellent mathematicians, and the comment that can be heard among mathematicians that econometricians are “self-made mathematicians” definitely does not apply to them. Therefore, given the task of writing an overview of asymptotic theory for minimization estimators for dependent processes, one could hardly imagine a better choice of writers for such a task. The result is a remarkable book that deserves to receive attention from an audience that is broader than experts in the field only.

The exact discrete model satisfied by equispaced data generated by a linear stochastic differential equations system is derived by a method that does not imply restrictions on observed discrete data per se. The method involves integrating the solution of the continuous time model in state space form and a nonstandard change in the order of three types of integration, facilitating the representation of the exact discrete model as an asymptotically time-invariant vector autoregressive moving average model. The method applying to the state space form is general and is illustrated using the prototypical higher order model for mixed stock and flow data discussed by Bergstrom (1986, Econometric Theory 2, 350–373).