Details
| Original language | English |
|---|---|
| Pages (from-to) | 189-200 |
| Number of pages | 12 |
| Journal | Computational Statistics and Data Analysis |
| Volume | 33 |
| Issue number | 2 |
| Publication status | Published - 28 Apr 2000 |
Abstract
In case of heteroscedasticity classical tests for trend in location do not control the type I error rate and have a reduced power. Since increasing locations are often accompanied by an increase in variability and apparent heteroscedasticity indicates that there are treatment differences, it is appropriate to consider an alternative involving trends in location as well as in scale. As a location-scale trend test we use the sum of a location and a scale trend test statistic. The investigated scale trend test is based on Levene's transformations. According to our simulation study we suggest to construct the tests using the principle of Bartholomew's likelihood ratio test. The different tests are applied to the data of a biomedical study.
Keywords
- Bartholomew's test, Heteroscedasticity, Levene's transformation, Location-scale trend test, Multiple contrast test
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
- Mathematics(all)
- Computational Mathematics
- Computer Science(all)
- Computational Theory and Mathematics
- Mathematics(all)
- Applied Mathematics
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Computational Statistics and Data Analysis, Vol. 33, No. 2, 28.04.2000, p. 189-200.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Parametric location-scale and scale trend tests based on Levene's transformation
AU - Neuhäuser, Markus
AU - Hothorn, Ludwig A.
PY - 2000/4/28
Y1 - 2000/4/28
N2 - In case of heteroscedasticity classical tests for trend in location do not control the type I error rate and have a reduced power. Since increasing locations are often accompanied by an increase in variability and apparent heteroscedasticity indicates that there are treatment differences, it is appropriate to consider an alternative involving trends in location as well as in scale. As a location-scale trend test we use the sum of a location and a scale trend test statistic. The investigated scale trend test is based on Levene's transformations. According to our simulation study we suggest to construct the tests using the principle of Bartholomew's likelihood ratio test. The different tests are applied to the data of a biomedical study.
AB - In case of heteroscedasticity classical tests for trend in location do not control the type I error rate and have a reduced power. Since increasing locations are often accompanied by an increase in variability and apparent heteroscedasticity indicates that there are treatment differences, it is appropriate to consider an alternative involving trends in location as well as in scale. As a location-scale trend test we use the sum of a location and a scale trend test statistic. The investigated scale trend test is based on Levene's transformations. According to our simulation study we suggest to construct the tests using the principle of Bartholomew's likelihood ratio test. The different tests are applied to the data of a biomedical study.
KW - Bartholomew's test
KW - Heteroscedasticity
KW - Levene's transformation
KW - Location-scale trend test
KW - Multiple contrast test
UR - http://www.scopus.com/inward/record.url?scp=0033750241&partnerID=8YFLogxK
U2 - 10.1016/S0167-9473(99)00051-1
DO - 10.1016/S0167-9473(99)00051-1
M3 - Article
AN - SCOPUS:0033750241
VL - 33
SP - 189
EP - 200
JO - Computational Statistics and Data Analysis
JF - Computational Statistics and Data Analysis
SN - 0167-9473
IS - 2
ER -