Robust testing for random effects in unbalanced heteroscedastic one-way models

Inkyung Jung and Pranab Kumar Sen

Journal of Nonparametric Statistics, 2008, vol. 20, issue 4, 305-317

Abstract: The usual variance ratio test for random effect, in a balanced design, is quite vulnerable to (i) unbalancedness, (ii) non-normality of either of the two random components, and (iii) heteroscedasticity of the chance errors. A robust rank-based test assuming only continuous, symmetric but otherwise arbitrary distributions for both the random effect and chance errors, and for a general heteroscedastic model is proposed here. Whereas the parametric tests are based on some F-distributional approximations, the proposed rank-based test rests on a normal approximation. Simulation studies, made to support the proposed methodology, suggest that not only the test is robust with respect to its significance level but also performs better in power, for heteroscedastic unbalanced models (even under normality).

Date: 2008
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/10485250802018477 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:gnstxx:v:20:y:2008:i:4:p:305-317

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/GNST20

DOI: 10.1080/10485250802018477

Access Statistics for this article

Journal of Nonparametric Statistics is currently edited by Jun Shao

More articles in Journal of Nonparametric Statistics from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().