Tests for Unit Roots and the Initial Condition

By Ulrich K. Müller, Graham Elliott. 2003

222 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/548cf280fa46e1eff2a2e4959bdb451b236f5f42

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

UNIT ROOT AND COINTEGRATING LIMIT THEORY WHEN INITIALIZATION IS IN THE INFINITE PAST (2008, unknown): “The importance of initial conditions in affecting size and power in inference has been particularly emphasized in recent work (Elliott, 1999; Müller and Elliott, 2003; Elliott and Müller, 2006; Harvey, Leybourne, and Taylor, 2009).”
Improving Size and Power in Unit Root Testing (2005, methodology): “The discussion emphasizes the role of nuisance parameters caused by the accommodation of serial correlation (Elliott, Rothenberg, and Stock, 1996), deterministic components (Elliott, Rothenberg, and Stock, 1996), and/or a nonzero initial condition (Müller and Elliott, 2003), but also touches upon tw”
Size and Power of Tests for Stationarity in Highly Autocorrelated Time Series (2005, unknown): “, for instance, Nabeya and Tanaka (1990) or Perron and Vodounou (2001)) that local-to-unity asymptotics provide much more accurate small sample approximations when a series contains a large autoregressive root than standard (jrjo1 fixed) asymptotics.”
Modelling income processes with lots of heterogeneity (2010, background): “This composite nature of the unit root hypothesis has been recognized in the time series literature (see, for example, Müller and Elliott (2003)), but there the main concern is how this a¤ects unit root tests.”
Testing for unit roots in bounded time series (2014, background): “Conversely, unit root tests based on standard critical values are 1As in Müller and Elliott (2003), we include the term T 1X̂20 in the numerator of MZ and MZt, so that the ADF (ADF t) and theMZ (MZt) statistics have the same limiting distributions. over-sized, with the degree of oversizing…”
Time‐Transformed Unit Root Tests for Models with Non‐Stationary Volatility (2008, methodology): “The lag truncation parameter, k, used in the computing the long-run variance estimator s(2)ARðkÞ was chosen according to MIC rule of Ng and Perron (2001) with k b12(T/100)c.”
Inflation Convergence and Divergence within the European Monetary Union (2012, unknown): “Monte Carlo experiments in Harvey and Bates (2003) and Busetti, Fabiani, and Harvey (2006) quantify the power properties of many unit-root tests for different initial conditions; the findings are in line with the arguments of Müller and Elliott (2003).”
The asymptotic size and power of the augmented Dickey–Fuller test for a unit root (2018, background): “The e ects of the initial value in unit root testing were de nitively investigated in the in uential article by Müller and Elliott (2003).”
Convergence of Prices and Rates of Inflation (2006, unknown): “…of the test will depend on the initial conditions, that is how far y 0 is from : If is known to be zero, the test based on the Dickey-Fuller (DF) tstatistic with no constant, denoted 0 ; is known to perform well, with a high value of jy 0 j actually enhancing power; see Müller and Elliott (2003).”
UNIT ROOT TESTING IN PRACTICE: DEALING WITH UNCERTAINTY OVER THE TREND AND INITIAL CONDITION (2009, unknown): “…PRACTICE 613 For the six unit root tests considered in Section 2, where we abstract from the trend issue for demeaned tests by setting β = 0, the following lemma details their asymptotic behavior, the proof of which follows directly from results in Phillips (1987b) and Müller and Elliott (2003).”