Details
Original language | English |
---|---|
Pages (from-to) | 793-800 |
Number of pages | 8 |
Journal | Biometrical journal |
Volume | 50 |
Issue number | 5 |
Publication status | Published - 17 Oct 2008 |
Abstract
This paper proposes a general approach for handling multiple contrast tests for normally distributed data in the presence of heteroscedasticity. Three candidate procedures are described and compared by simulations. Only the procedure with both comparison-specific degrees of freedom and a correlation matrix depending on sample variances maintains the α-level over all situations. Other approaches may fail notably as the variances differ more. Furthermore, related approximate simultaneous confidence intervals are given. The approach will be applied to a toxicological experiment.
Keywords
- Familywise error rate, Heteroscedasticity, Multiple contrast tests, Multivariate t-distribution, Simultaneous confidence intervals
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
- Decision Sciences(all)
- Statistics, Probability and Uncertainty
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Biometrical journal, Vol. 50, No. 5, 17.10.2008, p. 793-800.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Multiple contrast tests in the presence of heteroscedasticity
AU - Hasler, Mario
AU - Hothorn, Ludwig A.
PY - 2008/10/17
Y1 - 2008/10/17
N2 - This paper proposes a general approach for handling multiple contrast tests for normally distributed data in the presence of heteroscedasticity. Three candidate procedures are described and compared by simulations. Only the procedure with both comparison-specific degrees of freedom and a correlation matrix depending on sample variances maintains the α-level over all situations. Other approaches may fail notably as the variances differ more. Furthermore, related approximate simultaneous confidence intervals are given. The approach will be applied to a toxicological experiment.
AB - This paper proposes a general approach for handling multiple contrast tests for normally distributed data in the presence of heteroscedasticity. Three candidate procedures are described and compared by simulations. Only the procedure with both comparison-specific degrees of freedom and a correlation matrix depending on sample variances maintains the α-level over all situations. Other approaches may fail notably as the variances differ more. Furthermore, related approximate simultaneous confidence intervals are given. The approach will be applied to a toxicological experiment.
KW - Familywise error rate
KW - Heteroscedasticity
KW - Multiple contrast tests
KW - Multivariate t-distribution
KW - Simultaneous confidence intervals
UR - http://www.scopus.com/inward/record.url?scp=51649101445&partnerID=8YFLogxK
U2 - 10.1002/bimj.200710466
DO - 10.1002/bimj.200710466
M3 - Article
C2 - 18932141
AN - SCOPUS:51649101445
VL - 50
SP - 793
EP - 800
JO - Biometrical journal
JF - Biometrical journal
SN - 0323-3847
IS - 5
ER -