Bruce E. Hansen

Statistical Inference in Instrumental Variables Regression with I(1) Processes

This paper studies the asymptotic properties of instrumental variable (IV) estimates of multivariate cointegrating regressions and allows for deterministic and stochastic regressors as well as quite general deterministic processes in the data-generating mechanism. It is found that IV regressions are consistent even when the instruments are stochastically independent of the regressors. This phenomenon, which contrasts with traditional theory for stationary time series, is a beneficial artifact of spurious regression theory whereby stochastic trends in the instruments ensure their relevance asymptotically. Problems of inference are also addressed and some promising new theoretical results are reported. These involve a class of Wald tests which are modified by semiparametric corrections for serial correlation and for endogeneity. The resulting test statistics which we term fully-modified Wald tests have limiting X2 distributions, thereby removing the obstacles to inference in cointegrated systems that were presented by the nuisance parameter dependencies in earlier work. Some simulation results are reported which seek to explore the sampling behaviour of our suggested procedures. These simulations compare our fully modified (semiparametric) methods with the parametric error-correction methodology that has been extensively used in recent empirical research and with conventional least squares regression. Both the fully-modified and errorcorrection methods work well in finite samples and the sampling performance of each procedure confirms the relevance of asymptotic distribution theory, as distinct from super-consistency results, in discriminating between statistical methods.

Residual-based tests for cointegration in models with regime shifts

In this paper we examine tests for cointegration which allow for the possibility of regime shifts. We propose augmented Dickey-Fuller (ADF) and Phillips type tests designed to test the null of no cointegration against the alternative of cointegration in the presence of a possible regime shift. In particular we consider cases where the intercept and/or slope coefficients have a single break of unknown timing. A formal proof is provided for the limiting distributions of the various tests for the regime shift model (both a level and slope change). Critical values are calculated for the tests by simulation methods and a simple Monte Carlo experiment is conducted to evaluate finite sample performance. In the limited set of experiments, we find that the tests can detect cointegrating relations when there is a break in the intercept and/or slope coefficient. For these same experiments, the power of the conventional ADF test with no allowance for regime shifts falls sharply. As an illustration we test for structural breaks in the U.S. long-run money-demand equation using annual and quarterly data.

INFERENCE WHEN A NUISANCE PARAMETER IS NOT IDENTIFIED UNDER THE NULL HYPOTHESIS

Many econometric testing problems involve nuisance parameters which are not identified under the null hypotheses. This paper studies the asymptotic distribution theory for such tests. The asymptotic distributions of standard test statistics are described as functionals of chi-square processes. In general, the distributions depend upon a large number of unknown parameters. We show that a transformation based upon a conditional probability measure yields an asymptotic distribution free of nuisance parameters, and we show that this transformation can be easily approximated via simulation. The theory is applied to threshold models, with special attention given to the so-called self-exciting threshold autoregressive model. Monte Carlo methods are used to assess the finite sample distributions. The tests are applied to U.S. GNP growth rates, and we find that Potters (1995) threshold effect in this series can be possibly explained by sampling variation.

SAMPLE SPLITTING AND THRESHOLD ESTIMATION

Threshold models have a wide variety of applications in economics. Direct applications include models of separating and multiple equilibria. Other applications include empirical sample splitting when the sample split is based on a continuously-distributed variable such as firm size. In addition, threshold models may be used as a parsimonious strategy for non-parametric function estimation. For example, the threshold autoregressive model (TAR) is popular in the non-linear time series literature. Threshold models also emerge as special cases of more complex statistical frameworks, such as mixture models, switching models, Markov switching models, and smooth transition threshold models. It may be important to understand the statistical properties of threshold models as a preliminary step in the development of statistical tools to handle these more complicated structures. Despite the large number of potential applications, the statistical theory of threshold estimation is undeveloped. The previous literature has demonstrated that threshold estimates are super-consistent, but a distribution theory useful for testing and inference has yet to be provided. This paper develops a statistical theory for threshold estimation in the regression context. We allow for either cross-section or time series observations. Least squares estimation of the regression parameters is considered. An asymptotic distribution theory for the regression estimates (the threshold and the regression slopes) is developed. It is found that the distribution of the threshold estimate is non- standard. Methods to construct asymptotic confidence intervals are introduced, using both a t-statistic and LR-statistic approach. Monte Carlo simulations are presented to assess the accuracy of the asymptotic approximations. The empirical relevance of the theory is illustrated through an application to the multiple equilibria growth model of Durlauf and Johnson (1995).

Threshold effects in non-dynamic panels: Estimation, testing, and inference

Threshold regression methods are developed for non-dynamic panels with individual-specific fixed effects. Least squares estimation of the threshold and regression slopes is proposed using fixed-effects transformations. A non-standard asymptotic theory of inference is developed which allows construction of confidence intervals and testing of hypotheses. The methods are applied to a 15-year sample of 565 U.S. firms to test whether financial constraints affect investment decisions.

Autoregressive Conditional Density Estimation

R. F. Engles autoregressive conditional heteroskedastic model is extended to permit parametric specifications for conditional dependence beyond the mean and variance. The suggestion is to model the conditional density with a small number of parameters, and then model these parameters as functions of the conditioning information. This method is applied to two data sets. The first application is to the monthly excess holding yield on U.S. Treasury securities, where the conditional density used is a Students t distribution. The second application is to the U.S. Dollar/Swiss Franc exchange rate, using a new skewed Student t conditional distribution. Copyright 1994 by Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association.

Tests for Parameter Instability in Regressions With I(1) Processes

This article derives the large-sample distributions of Lagrange multiplier (LM) tests for parameter instability against several alternatives of interest in the context of cointegrated regression models. The fully modified estimator of Phillips and Hansen is extended to cover general models with stochastic and deterministic trends. The test statistics considered include the SupF test of Quandt, as well as the LM tests of Nyblom and of Nabeya and Tanaka. It is found that the asymptotic distributions depend on the nature of the regressor processes—that is, if the regressors are stochastic or deterministic trends. The distributions are noticeably different from the distributions when the data are weakly dependent. It is also found that the lack of cointegration is a special case of the alternative hypothesis considered (an unstable intercept), so the tests proposed here may also be viewed as a test of the null of cointegration against the alternative of no cointegration. The tests are applied to three data se...

Erratum: The Likelihood Ratio Test under Nonstandard Conditions: Testing the Markov Switching Model of GNP

A theory of testing under non-standard conditions is developed. By viewing the likelihood as a function of the unknown parameters, empirical process theory enables us to bound the asymptotic distribution of standardized likelihood ratio statistics, even when conventional regularity conditions (such as unidentified nuisance parameters and identically zero scores) are violated. This testing methodology is applied.to the Markov switching model of GNP proposed by Hamilton (1989). The standardized likelihood ratio test is unable to reject the hypothesis of an AR(4) in favour of the Markov switching model. Instead, we find strong evidence for an alternative model. This model, like Hamiltons, is characterized by parameters which switch between states, but the states arrive independently over time, rather than following an unrestricted Markov process. The primary difference, however, is that the second autoregressive parameter, in addition to the intercept, switches between states. Copyright 1992 by John Wiley & Sons, Ltd.

Testing for two-regime threshold cointegration in vector error-correction models

This paper examines a two-regime vector error-correction model with a single cointegrating vector and a threshold effect in the error-correction term. We propose a relatively simple algorithm to obtain maximum likelihood estimation of the complete threshold cointegration model for the bivariate case. We propose a SupLM test for the presence of a threshold. We derive the null asymptotic distribution, show how to simulate asymptotic critical values, and present a bootstrap approximation. We investigate the performance of the test using Monte Carlo simulation, and find that the test works quite well. Applying our methods to the term structure model of interest rates, we find strong evidence for a threshold effect.

Testing for parameter instability in linear models

Simple tests for parameter instability are presented and discussed. These tests have locally optimal power and do not require a priori knowledge of “the breakpoint.” Two empirical examples are presented to illustrate the use of the tests. The first examines whether an AR(1) model for annual U.S. output growth rates has remained stable over 1889–1987. The second examines the stability of an error correction model for an aggregate life cycle model of consumption.