Testing Weak Cross-Sectional Dependence in Large Panels

By M. Hashem, Pesaran, George Kapetanios, Ron P. Smith, T. Yamagata, Aman Ullah. Social Science Research Network, 2012

2,633 citations recorded by the source corpus.

This article considers testing the hypothesis that errors in a panel data model are weakly cross-sectionally dependent, using the exponent of cross-sectional dependence α, introduced recently in Bailey, Kapetanios, and Pesaran (2012). It is shown that the implicit null of the cross-sectional dependence (CD) test depends on the relative expansion rates of N and T. When T = O(N ε), for some 0 < ε ≤1, then the implicit null of the CD test is given by 0 ≤ α < (2 − ε)/4, which gives 0 ≤ α <1/4, when N and T tend to infinity at the same rate such that T/N → κ, with κ being a finite positive constant. It is argued that in the case of large N panels, the null of weak dependence is more appropriate than the null of independence which could be quite restrictive for large panels. Using Monte Carlo experiments, it is shown that the CD test has the correct size for values of α in the range [0, 1/4], for all combinations of N and T, and irrespective of whether the panel contains lagged values of the dependent variables, so long as there are no major asymmetries in the error distribution.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/ec910f8217d0555e691863ffc5b5470384ca70e9

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

Theoretical Framework for the Carbon Emissions Effects of Technological Progress and Renewable Energy Consumption (2021, methodology): “The issue of cross-section dependence is dealt with using a newly proposed approach by Chudik et al. (2017) and Chudik and Pesaran (2015).”
Estimation and inference for spatial models with heterogeneous coefficients: An application to US house prices (2020, positive): “Aquaro, M., N. Bailey, and M. H. Pesaran (2015).”
Interplay between technological innovation and environmental quality: Formulating the SDG policies for next 11 economies (2020, methodology): “In the empirical analysis, as a first step, we have checked the cross-sectional dependence in the 1 data by using the weak cross-sectional dependence test devised by Chudik and Pesaran (2015).”
Audit Committee, Corporate Governance and Firm Performance: Empirical Evidence from India (2016, methodology): “Symbolically, the hypothesis could be expressed as: H0: E(uit, ujt) = 0,for all t and i ≠ j (Pesaran, 2015) For checking the heteroskedasticity in the residuals of cross sections, white test was applied with Huber white coefficient covariance estimator on year wise cross section data.”
Testing for Panel Cointegration Using Common Correlated Effects Estimators (2017, methodology): “The computation of the CD test statistic in Pesaran (2015) leads us to reject the null hypothesis of no cross section correlation for the panels of the variables involved in the model, which indicates that panel cointegration test statistics that account for the presence of cross section dependence…”
The Volatility of Capital Flows in Emerging Markets: Measures and Determinants (2017, methodology): “33 In the former case, we use the Durbin-Wu-Hausman test, while in the latter we perform the test for weak crosssectional dependence proposed by Pesaran (2015).”
Exponent of Cross-Sectional Dependence: Estimation and Inference (2016, positive): “Pesaran (2015) shows that when T = O(Nd) for some 0 < d ≤ 1, then the implicit null of the CD test is given by 0 ≤ α < (2−d)/4, and it is asymptotically distributed as N(0, 1).”
Testing Cross-Sectional Correlation in Large Panel Data Models with Serial Correlation (2016, methodology): “For the weak error cross-sectional correlation case, we conjecture that it is similar to Pesaran [15].”
Public debt and economic growth: panel data evidence for Asian countries (2020, methodology): “…0.018 0.016 Adjusted R2 0.408 0.512 0.700 0.828 Number of observations 448 448 448 448 Note: the estimation includes the dummy for the global financial crisis and constant Number in parentheses is the standard error ***, **, * shows 1%, 5% and 10% of significance levels Chudik and Pesaran (2015).”