Zacharias Psaradakis

Detecting Periodically Collapsing Bubbles: A Markov-Switching Unit Root Test

This paper addresses the problem of testing for the presence of a stochastic bubble in a time series in the case that the bubble is periodically collapsing so that the asset price keeps returning to the level implied by the market fundamentals. As this is essentially a problem of identifying the collapsing periods from the expanding ones, we propose using a generalization of the Dickey-Fuller test procedure which makes use of the class of Markov regime-switching models. The potential of the new methodology is illustrated via simulation, and an empirical example is given.

ON THE DETERMINATION OF THE NUMBER OF REGIMES IN MARKOV‐SWITCHING AUTOREGRESSIVE MODELS

Dynamic models with parameters that are allowed to depend on the state of a hidden Markov chain have become a popular tool for modelling time series subject to changes in regime. An important question that arises in applications involving such models is how to determine the number of states required for the model to be an adequate characterization of the observed data. In this paper, we investigate the properties of alternative procedures that can be used to determine the state dimension of a Markov-switching autoregressive model. These include procedures that exploit the ARMA representation which Markov-switching processes admit, as well as procedures that are based on optimization of complexity-penalized likelihood measures. Our Monte Carlo analysis reveals that such procedures estimate the state dimension correctly, provided that the parameter changes are not too small and the hidden Markov chain is fairly persistent. The use of the various methods is also illustrated by means of empirical examples. Copyright 2003 Blackwell Publishing Ltd.

COINTEGRATION AND CHANGES IN REGIME: THE JAPANESE CONSUMPTION FUNCTION

In this paper we examine a model of cointegration where long-run parameters are subject to switching between several different cointegrating regimes. These shifts are allowed to be governed by the outcome of an unobserved Markov chain with unknown transition probabilities. We illustrate this approach using Japanese data on consumption and disposable income, and find that the data favor a Markov-switching long-run relationship over a standard temporally stable formulation.

Switching error-correction models of house prices in the United Kingdom☆

Abstract This paper develops error-correction models of real house prices in the UK in which the adjustment coefficient switches stochastically between a stable regime where disequilibrium correction takes place and an unstable regime where such a correction does not occur. The generating mechanism of the shifts is modelled as a Markov process with transition probabilities which are either time-invariant or depend on the extent to which the system is out of disequilibrium. Estimation of error-correction models for the UK reveals that the observed booms in real house prices are associated with an unstable regime. We also find that the probability that the system remains in an unstable regime decreases as deviations from equilibrium increase.

Finite-sample properties of the maximum likelihood estimator in autoregressive models with Markov switching

This paper examines the finite-sample properties of the maximum likelihood estimator in autoregressive models subject to Markov mean and variance shifts. Our results reveal that conventional asymptotic approximations to the distribution of the maximum likelihood estimator can often be poor for the sample sizes that are typical for annual and quarterly times series.

Joint Determination of the State Dimension and Autoregressive Order for Models with Markov Regime Switching

This paper is concerned with the problem of joint determination of the state dimension and autoregressive order of models with Markov-switching parameters. A model selection procedure is proposed which is based on optimization of complexity-penalized likelihood criteria. The efficacy of the procedure is evaluated by means of Monte Carlo experiments. Copyright 2006 The Authors Journal compilation 2006 Blackwell Publishing Ltd.

Bootstrap Tests for an Autoregressive Unit Root in the Presence of Weakly Dependent Errors

This paper examines bootstrap tests of the null hypothesis of an autoregressive unit root in models that may include a linear rend and/or an intercept and which are driven by innovations that belong to the class of stationary and invertible linear processes. Our approach makes use of a sieve bootstrap procedure based on residual resampling from autoregressive approximations, the order of which increases with the sample size at a suitable rate. We show that the sieve bootstrap provides asymptotically valid tests of the unit‐root hypothesis and demonstrate the small‐sample effectiveness of the method by means of simulation.

A simple procedure for detecting periodically collapsing rational bubbles

This paper proposes a new procedure for detecting the presence of periodically collapsing rational bubbles via an analysis of the properties of the relevant observable time series. The procedure is based on random coefficient autoregressive models. An empirical application of the procedure to German hyperinflation data is examined and discussed.

On the power of tests for superexogeneity and structural invariance

Abstract This paper examines the finite-sample power of tests of structural invariance and superexogeneity hypotheses in econometric models with contemporaneous conditioning variables. We consider both direct parametric tests of superexogeneity, as well as indirect procedures based on temporal stability tests for the parameters of interest. Our Monte Carlo analysis reveals that both types of tests may lack power in interesting classes of models. An empirical illustration investigates the superexogeneity of the short-term interest rate in a dynamic specification for the U.S. term structure.

A Reconciliation of Some Paradoxical Empirical Results on the Expectations Model of the Term Structure

In this paper, the authors attempt to reconcile contradictory empirical results for the expectations model of the term structure which are found when it is tested by a variety of methods based on single-equation and vector autoregression (VAR) models. Using monthly data for one-month and three-month interest rates, the authors show that the expectations hypothesis is rejected for the United States and United Kingdom term structures on the basis of some popular tests. However, tests based on VAR models or on instrumental variables regressions of yield spreads on future short rate changes provide no evidence against the expectations model with a random component in the term premium. Copyright 1997 by Blackwell Publishing Ltd