Nikitas Pittis

Causality and forecasting in incomplete systems

In this paper we examine how causality inference and forecasting within a bivariate VAR, consisting of y(t) and x(t), are affected by the omission of a third variable, w(t), which causes (a) none, (b) one, and (c) both variables in the bivariate system. We also derive conditions under which causality inference and forecasting are invariant to the selection of a bivariate or a trivariate model. The most general condition for the invariance of both causality and forecasting to model selection is shown to require the omitted variable not to cause any of the variables in the bivariate system, although it allows the omitted variable to be caused by the other two. We also show that the conditions for one-way causality inference to be invariant to model selection are not sufficient to ensure that forecasting will also be invariant to the model selected. Finally, we present a numerical illustration of the potential losses, in terms of the variance of the forecast, as a function of the forecast horizon and for alternative parameter values—they can be rather large, as the omission of a variable can make the incomplete model unstable.

Volatility transmission and financial crises

In this paper we examine the international transmission of the 1997 South East Asia financial crisis. We estimate a bivariate GARCH-BEKK model, and carry out LR tests for causality-in-variance with bootstrapped critical values. Three pairwise models are estimated for US, European, Japanese and South East Asian daily stock market returns. Volatility spillovers are found in all cases. The dynamics of the conditional volatilities differ, but causality links in the variance are found to be strong and bidirectional in normal periods, and unidirectional (from the markets in turmoil to the others) following the onset of the crisis, consistently with crisiscontingent models.

Common stochastic trends and inflation convergence in the EMS

Common Stochastic Trends and Inflation Convergence in the EMS. — This article seeks to clarify the relationship between the concept of convergence and univariate and cointegration analysis by looking at inflation convergence within the EMS. We take issue with the view put forward by Artis and Nachane, who hold that cointegration of inflation in Germany with inflation in the other EMS countries is a necessary condition for the “german leadership” thesis to stand. We think that, on the contrary, as long as convergence is still in the process of being achieved, inflation differentials are likely to be non-stationary and, if so, to exhibit common stochastic trends. However, our empirical results, based on Phillips-Perron tests and on the maximum likelihood framework developed by Johansen to test for common trends and cointegration, confirm the validity of the German leadership hypothesis.ZusammenfassungGemeinsame stochastische Trends und Konvergenz der Inflation im EWS. — In diesem Aufsatz soll die Beziehung zwischen dem Konzept der Konvergenz und univariater Kointegrations-Analyse geklärt werden, indem die Inflationskonvergenz im EWS betrachtet wird. Ausgangspunkt ist die Ansicht von Artis und Nachane, die die Kointegration der Inflation in Deutschland mit der in den anderen EWS-Ländern als eine notwendige Bedingung für die These von der “deutschen Führung” ansehen. Dagegen wird hier die Ansicht vertreten, daβ - solange sich die Konvergenz noch herausbildet - die Inflationsunterschiede eher nichtstationär sind und demnach gemeinsame stochastische Trends aufweisen. Die empirischen Ergebnisse, die auf Phillips-Perron-Tests und auf dem von Johansen im Rahmen der Methode der gröβten Dichte entwickelten Verfahren zur Prüfung von gemeinsamen Trends und Kointegration beruhen, bestätigen allerdings die Hypothese der deutschen Führungsrolle.

Cointegration and joint efficiency of international commodity markets

Abstract This paper investigates the semi-strong efficiency hypothesis in the international commodity markets of four industrialized countries, using vector autoregression (VAR) and cointegration techniques. Efficiency in these markets requires the corresponding real exchange rates to be martingales with respect to any information set available in the public domain. In the context of a VAR consisting only of real exchange rates, we show that necessary and sufficient conditions for joint efficiency of all the markets under consideration amount to the VAR being of order one (Markovness) and non-cointegrated. On the contrary, in a VAR extended by other potentially “relevant” variables, such as the corresponding real interest rates, non-cointegration and Markovness are only sufficient conditions for the same commodity markets to be characterized as jointly efficient. We also suggest methods for efficiency testing in each individual market within a cointegrated VAR and, finally, we discuss possible long-run linkages among the real exchange rates and real interest rates in association with efficiency in the commodity markets. JEL Classification Number: F31

Causality Links between Consumer and Producer Prices: Some Empirical Evidence

This paper reexamines the relationship between consumer and producer prices in the G7 countries (United States, Canada, Germany, France, Italy, United Kingdom, and Japan), and it improves on the existing literature in two ways. First, it takes into account causality links arising from the transmission mechanism of monetary policy, which are generally overlooked. Second, it employs the causality testing method for unstable systems recently introduced by, which results in standard asymptotics, thereby obtaining valid statistical inference. The empirical results are consistent with the conventional wisdom according to which there is unidirectional causality running from producer to consumer prices, bidirectional causality (or even no significant links) only being found when the causality links reflecting the monetary transmission mechanism are ignored.

Estimator Choice and Fisher's Paradox: A Monte Carlo Study

Abstract This paper argues that Fishers paradox can be explained away in terms of estimator choice. We analyse by means of Monte Carlo experiments the small sample properties of a large set of estimators (including virtually all available single-equation estimators), and compute the critical values based on the empirical distributions of the t-statistics, for a variety of Data Generation Processes (DGPs), allowing for structural breaks, ARCH effects etc. We show that precisely the estimators most commonly used in the literature, namely OLS, Dynamic OLS (DOLS) and non-prewhitened FMLS, have the worst performance in small samples, and produce rejections of the Fisher hypothesis. If one employs the estimators with the most desirable properties (i.e., the smallest downward bias and the minimum shift in the distribution of the associated t-statistics), or if one uses the empirical critical values, the evidence based on US data is strongly supportive of the Fisher relation, consistently with many theoretical models.

UNIT ROOT TESTING USING COVARIATES: SOME THEORY AND EVIDENCE{

This paper analyzes the conditions under which power gains can be achieved using the Covariate Augmented Dickey-Fuller test (CADF) rather than the conventional Augmented Dickey-Fuller (ADF), and argues that this method has the advantage, relative to univariate unit root tests, of increasing power without suffering from the large size distortions affecting the latter. The inclusion of covariates affects unit root testing by: (a) reducing the standard error of the estimate of the autoregressive parameter without affecting the estimate itself, and/or (b) reducing both the standard error and the absolute value of the estimate itself. Conditions in terms of contemporaneous correlation and Granger causality are derived for case (a) or (b) to arise. As an illustration, it is shown that applying the more powerful CADF (rather than the ADF) test reverses the finding of a unit root for many US macroeconomic series. Copyright 1999 by Blackwell Publishing Ltd

Testing for PPP: the erratic behaviour of unit root tests

Abstract In this article we argue that the type of stationarity exhibited by the real exchange rate cannot be accommodated by the fixed-parameter autoregressive homoscedastic models normally employed in the literature. Using a dataset including 39 countries and spanning a period of up to two centuries, we analyse the behaviour of both WPI- and CPI-based measures of the real exchange rate. In particular, we compute a recursive t -statistic, and show that it has an erratic behaviour, suggesting the presence of endemic instability. It appears that the type of non-stationarity present in the data is more complex than the unit root one. This explains the contradictory results of previous studies using standard unit root tests for establishing whether PPP holds.

Cointegration and predictability of asset prices1

Abstract This article argues that whatever concerns one might have about the identification of a cointegrating relationship with market inefficiency, cointegration tests can still be usefully employed to investigate the predictability of asset prices. We examine the case of n -dimensional systems, and show that the standard assumption made in the literature that cointegration implies predictability of all n asset prices is not valid. In the presence of r cointegrating vectors, only r prices are predictable, and standard Wald or LM tests carried out within Johansens Maximum Likelihood (ML) framework can be used to establish for which prices unpredictability does not hold. Similarly, one can distinguish between (joint) unpredictability in the short- and long-run when asset prices are driven by fundamentals.

Testing for PPP and UIP in an FIML framework: Some evidence for Germany and Japan

Abstract This paper argues that international parity conditions should be tested using an econometric approach, which allows for possible interactions in the determination of prices, interest rates, and exchange rates, and also for different short- and long-run dynamics. Failure to do so might account for the apparent lack of empirical success of exchange rate models based on the purchasing power parity (PPP) and uncovered interest rate parity (UIP) hypotheses. By focusing on the German mark and the Japanese yen (both nominal effective rates and bilateral rates vis-a-vis the U.S. dollar), we show that empirical support for PPP and UIP can be found within a full-information maximum-likelihood (FIML) framework. This confirms that such structural hypotheses are to be tested in a multivariate cointegration system in order to draw valid inference.