This article discusses the properties of univariate Dickey–Fuller test and Johansen for cointegrating rank when there exist additive outlying observations in time series. We provide analytical as well numerical evidence that outliers may produce spurious stationary. Hence will reject a unit root too frequently indicate many vectors. The results easily generalize to models with "temporary change" outliers. Through an empirical example we discuss how temporary change can be detected practice,...

This paper provides a selective survey of the recent literature dealing with I(2) variables in economic time series, that is, processes require to be differenced twice order become stationary. With reference particular models intuition is provided why I(2)‐and polynomial cointegration are features likely occur economics. The properties series discussed and I review topics such as: Testing for double unit roots, representations cointegrated systems, hypothesis testing single equations as well...

EU labour markets Effects of greater product market integration European are in a state flux due to the changing situation induced by international integration. This process affects wage formation through more fierce competition and increased mobility jobs. development is some observers taken enforce flexibility, while for others it signals an erosion social standards turn possibly welfare society. Since not very mobile Europe, effects integration on mostly indirect via We review channels...

textabstractThe properties of the univariate Dickey-Fuller test and Johansen for cointegrating rank when there exist additive outlying observations in time series are examined. The analysis provides analytical as well numerical evidence that outliers may produce spurious stationarity. Hence, will reject a unit root too frequently, indicate many vectors. results easily generalize to models with temporary change outliers. Through an empirical example, demonstrates how can be detected practice,...

Multicointegration, in the sense of Granger and Lee (1990), frequently occurs models stock‐flow adjustment implies cointegration amongst I(2) variables their differences (polynomial cointegration). The purpose this article is two‐fold. First, we demonstrate that based on a multicointegrated vector autoregression (VAR) two equivalent error correction model (ECM) representations can be derived; first expressed terms adjustments flows (the standard ECM), second both stocks flows. Secondly,...

The notion of separation in cointegrated systems helps identifying possible sub‐system structures that may reduce the complexity larger by yielding a more parsimonious representation times series. In this paper we demonstrate although subsystem cointegration analysis such can be conducted case both completely and partially separated systems, dual approach, i.e. calculation common stochastic trends, turn out to yield properties trends differ depending upon type under consideration....

In this paper we demonstrate a new way of testing the linear quadratic adjustment cost (LQAC) model under rational expectations. We illustrate how parameter restrictions arising from can be formally specified and use these to extent technique Campbell Shiller (1987) wider class models based on present value relations. Potentially demand for labour is an area in which LQAC find applicability practice subsequently analyse sectoral Danish manufacturing. find, however, that our data set...

The functioning of electricity markets has experienced increasing complexity as a result deregulation in recent years. Consequently this affects the multilateral price behaviour across regions with physical exchange power. It been documented elsewhere that features such long memory and regime switching reflecting congestion non-congestion periods are empirically relevant hence need to be taken into account when modeling behavior. In present paper we further elaborate on co-existence switches...

The Lee and Carter (1992) model assumes that the deterministic stochastic time series dynamics load with identical weights when describing development of age-specific mortality rates. Effectively this means main characteristics simplify to a random walk drift components. But restricting adjustment mechanism linear trend components be may too strong simplification. In fact, presence component itself result from bias induced by properly fitting characterizes data. We find empirical evidence...

Abstract In this article, I argue that if an explosive root is considered a serious alternative to 1(2) process, then joint testing for the number of unit roots along lines Hasza and Fuller preferable instead prior differencing as suggested by Dickey Pantula. A semiparametric equivalent Hasza–Fuller test permits rather general assumptions about innovation process developed. The straightforward generalization Phillips Z single root. Through Monte Carlo simulations, different tests are...

A frequent criticism of unit root tests concerns the poor power and size properties that many such exhibit. However, past decade or so intensive research has been conducted to alleviate these problems great advances have made. The present paper provides a selective survey recent contributions improve upon both in doing approach using rigorous statistical optimality criteria development is stressed. In addition presenting where improved can be achieved by modifying standard Dickey-Fuller...