Testing for stationarity in heterogeneous panel data where the time dimension is finite

By K. Hadri, Rolf Larsson. 2005

117 citations recorded by the source corpus.

and hence makes the test valid for any (T, N) combination. The asymptotic distributions of the tests are derived under the null and are shown to be normally distributed. Their moments for T fixed are derived analytically using Ghazal's (1994, Statistics and Probability letters 20, 313--319) 1 lemma 1. Finite sample size and power are considered in a Monte Carlo experiment. The proposed tests have empirical sizes that are very close to the nominal 5% level. The Monte Carlo results clearly show that the power of the test statistics increases substantially with N, T and ω (ω being the number of unit root processes under the alternative). The results indicate that the assumption that T is asymptotic rather than fixed leads to tests that are substantially oversized particularly for relatively short panels with large N. Copyright 2005 Royal Economic Society

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/70e7ca844e128817ce93b47c854712709d7f5026

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

Citations in the corpus

Unit Roots and Cointegration in Panels (2005, methodology): “The individual test statistics can be combined as in the test suggested by IPS (2003) yielding κ̄ = N−1/2 ∑N i=1 [κi − E(κi)]√ V ar(κi) , where asymptotic values of E(κi) and V ar(κi) are derived in Hadri (2000) and values for finite T and N → ∞ are presented in Hadri and Larsson (2005).”
Ownership Level, Ownership Concentration and Liquidity (2007, unknown): “6 Also, for short panels where T (number of periods) is relatively small to N (number of firms), the tests for stationarity developed under the assumption of large T are usually unsatisfactory because the rate condition N/T 0 limits the application of the tests to cases where T is larger than N (Had”
The Performance of Panel Unit Root and Stationarity Tests: Results from a Large Scale Simulation Study (2006, methodology): “Figure 7: Power of the Hadri (2000) test (upper row) and of the Hadri and Larsson (2002) test (lower row) for case 2 (DGP2(0, 1 ,c )) against the alternative of a unit root for c ∈ {−0.2, −0.4, −0.8, −0.9, −0.95} and T ∈{ 10, 25, 100}.”
Towards the Estimation of Equilibrium Exchange Rates for CEE Acceding Countries: Methodological Issues and a Panel Cointegration Perspective (2004, background): “The Hadri-test accounts for the presence of autocorrelated errors and employs the finite sample critical values following Hadri and Larsson (2002) 24 Kim and Korhonen (2002) and DeBroek and Sloek (2001) find strong evidence for a significantly negative impact of this variable on the real exchange…”
Panel Stationarity Test with Structural Breaks (2008, unknown): “As in Hadri and Larsson (2005), it is easy to show, using simulations , that the power of the tests increases as the proportion of unit roots increases under the alternative.”
Panel Unit Root Tests in the Presence of Cross-Sectional Dependence: Finite Sample Performance and an Application (2009, unknown): “(2002), Im, Pesaran and Shin (2003) and Hadri and Larsson (2005) assume the independence of individual”
The Impact of Earnings on the Pricing of Credit Default Swaps (2007, methodology): “The Hadri and Larsson (2005) test statistic for panel data is robust to unbalanced panels and hetreoscedastic errors.”