We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

econ.EM

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Economics > Econometrics

Title: A Bias-Corrected CD Test for Error Cross-Sectional Dependence in Panel Data Models with Latent Factors

Authors: M. Hashem Pesaran (1 and 2), Yimeng Xie (3) ((1) University of Southern California, USA, (2) Trinity College, Cambridge, UK, (3) Xiamen University, China)
Abstract: In a recent paper Juodis and Reese (2021) (JR) show that the application of the CD test proposed by Pesaran (2004) to residuals from panels with latent factors results in over-rejection and propose a randomized test statistic to correct for over-rejection, and add a screening component to achieve power. This paper considers the same problem but from a different perspective and shows that the standard CD test remains valid if the latent factors are weak, and proposes a simple bias-corrected CD test, labelled CD*, which is shown to be asymptotically normal, irrespective of whether the latent factors are weak or strong. This result is shown to hold for pure latent factor models as well as for panel regressions with latent factors. Small sample properties of the CD* test are investigated by Monte Carlo experiments and are shown to have the correct size and satisfactory power for both Gaussian and non-Gaussian errors. In contrast, it is found that JR's test tends to over-reject in the case of panels with non-Gaussian errors, and have low power against spatial network alternatives. The use of the CD* test is illustrated with two empirical applications from the literature.
Subjects: Econometrics (econ.EM)
Cite as: arXiv:2109.00408 [econ.EM]
  (or arXiv:2109.00408v1 [econ.EM] for this version)

Submission history

From: Yimeng Xie [view email]
[v1] Wed, 1 Sep 2021 14:36:30 GMT (55kb)

Link back to: arXiv, form interface, contact.