Markku Lanne

Comparison of unit root tests for time series with level shifts

Unit root tests are considered for time series which have a level shift at a known point in time. The shift can have a very general nonlinear form, and additional deterministic mean and trend terms are allowed for. Prior to the tests, the deterministic parts and other nuisance parameters of the data generation process are estimated in a first step. Then, the series are adjusted for these terms and unit root tests of the Dickey–Fuller type are applied to the adjusted series. The properties of previously suggested tests of this sort are analysed and modifications are proposed which take into account estimation errors in the nuisance parameters. An important result is that estimation under the null hypothesis is preferable to estimation under local alternatives. This contrasts with results obtained by other authors for time series without level shifts.

Test Procedures for Unit Roots in Time Series with Level Shifts at Unknown Time

Two types of unit root tests which accommodate a structural level shift at a known point in time are extended to the situation where the break date is unknown. It is shown that for any estimator for the break date the tests have the same asymptotic distribution as the corresponding tests under the known break date assumption. Different estimators of the break date are compared in a Monte Carlo experiment and a recommendation for choosing the break date in small samples is given. It is also shown that ignoring the fact that a break has occurred and applying a standard unit root test may lead to substantial size distortion and total loss of power. Example series from the Nelson-Plosser data set are used to illustrate the performance of our tests.

TESTING THE PREDICTABILITY OF STOCK RETURNS

Previous literature indicates that stock returns are predictable by several strongly autocorrelated forecasting variables, especially at longer horizons. It is suggested that this finding is spurious and follows from a neglected near unit root problem. Instead of the commonly used t-test, we propose a test that can be considered as a general test of whether the return can be predicted by any highly persistent variable. Using this test, no predictability is found for U.S. stock return data from the period 1928-1996. Simulation experiments show that the standard t-test clearly overrejects whereas our proposed test controls size much better.

Structural Vector Autoregressions with Nonnormal Residuals

In structural vector autoregressive (SVAR) modeling, sometimes the identifying restrictions are insufficient for a unique specification of all shocks. In this paper it is pointed out that specific distributional assumptions can help in identifying the structural shocks. In particular, a mixture of normal distributions is considered as a possible model that can be used in this context. Our model setup enables us to test restrictions which are just-identifying in a standard SVAR framework. The results are illustrated using a U.S. macro data set and a system of U.S. and European interest rates.

Nonlinear Dynamics of Interest Rate and Inflation

According to several empirical studies US inflation and nominal interest rates as well as the real interest rate can be described as unit root processes. These results imply that nominal interest rates and expected inflation do not move one-for-one in the long run, which is incongruent with theoretical models. In this paper we introduce a new nonlinear bivariate mixture autoregressive model that seems to fit quarterly US data (1953 : II-2004 : IV) reasonably well. It is found that the three-month Treasury bill rate and inflation share a common nonlinear component that explains a large part of their persistence. The real interest rate is devoid of this component, indicating one-for-one movement of the nominal interest rate and inflation in the long run and, hence, stationarity of the real interest rate. Copyright

Near unit roots, cointegration, and the term structure of interest rates

The term structure of interest rates is often modelled as a cointegrated system with the yield spreads forming the cointegrating vectors. Testing whether the yield spreads span the cointegration space is problematic because conventional tests on the cointegration vectors tend to overreject when the largest autoregressive roots deviate from unity, as is likely to be the case with interest rates. A new test that is robust w.r.t. deviations from the exact unit root assumption is developed and applied to monthly US interest rate data from 1952:1-1991:2. Taking into account the regime shift in 1979, the hypothesis of the yield spreads being the cointegrating vectors cannot be rejected using the robust test. Copyright

Noncausal Autoregressions for Economic Time Series

This paper is concerned with univariate noncausal autoregressive models and their potential usefulness in economic applications. In these models, future errors are predictable, indicating that they can be used to empirically approach rational expectations models with nonfundamental solutions. In the previous theoretical literature, nonfundamental solutions have typically been represented by noninvertible moving average models. However, noncausal autoregressive and noninvertible moving average models closely approximate each other, and therefore, the former provide a viable and practically convenient alternative. We show how the parameters of a noncausal autoregressive model can be estimated by the method of maximum likelihood and derive related test procedures. Because noncausal autoregressive models cannot be distinguished from conventional causal autoregressive models by second order properties or Gaussian likelihood, a model selection procedure is proposed. As an empirical application, we consider modeling the U.S. inflation which, according to our results, exhibits purely forward-looking dynamics.

NONCAUSAL VECTOR AUTOREGRESSION

In this paper, we propose a new noncausal vector autoregressive (VAR) model for non-Gaussian time series. The assumption of non-Gaussianity is needed for reasons of identifiability. Assuming that the error distribution belongs to a fairly general class of elliptical distributions, we develop an asymptotic theory of maximum likelihood estimation and statistical inference. We argue that allowing for noncausality is of importance in empirical economic research, which currently uses only conventional causal VAR models. Indeed, if noncausality is incorrectly ignored, the use of a causal VAR model may yield suboptimal forecasts and misleading economic interpretations. This is emphasized in the paper by noting that noncausality is closely related to the notion of nonfundamentalness, under which structural economic shocks cannot be recovered from an estimated causal VAR model. As detecting nonfundamentalness is therefore of great importance, we propose a procedure for discriminating between causality and noncausality that can be seen as a test of nonfundamentalness. The methods are illustrated with applications to fiscal foresight and the term structure of interest rates.

Unit root tests for time series with level shifts: a comparison of different proposals

A number of unit root tests which accommodate a deterministic level shift at a known point in time are compared in a Monte Carlo study. The tests differ in the way they treat the deterministic term of the DGP. It turns out that Phillips-Perron type tests have very poor small sample properties and cannot be recommended for applied work. Moreover, tests which estimate the deterministic term by a GLS procedure under the unit root null hypothesis are superior in terms of size and power properties relative to tests which estimate the deterministic term by OLS procedures.

Modeling Conditional Skewness in Stock Returns

Abstract In this paper, we propose a new GARCH-in-Mean (GARCH-M) model allowing for conditional skewness. The model is based on the so-called z distribution capable of modeling skewness and kurtosis of the size typically encountered in stock return series. The need to allow for skewness can also be readily tested. The model is consistent with the volatility feedback effect in that conditional skewness is dependent on conditional variance. Compared to previously presented GARCH models allowing for conditional skewness, the model is analytically tractable, parsimonious and facilitates straightforward interpretation.Our empirical results indicate the presence of conditional skewness in the monthly postwar US stock returns. Small positive news is also found to have a smaller impact on conditional variance than no news at all. Moreover, the symmetric GARCH-M model not allowing for conditional skewness is found to systematically overpredict conditional variance and average excess returns.