Zusammenfassung
In dieser Arbeit werden einige neue Methoden aus der Biometrie vorgestellt. Zunächst
wird der Vergleich zweier Gruppen behandelt. Neben dem Wilcoxon-Rangsummentest werden
zwei Alternativen, der Baumgartner-Weiß-Schindler-Test sowie der Brunner-Munzel-Test,
vorgestellt. Insbesondere wird die Situation ungleicher Variabilitäten besprochen.
Das Thema des zweiten Abschnitts sind Kombinationsmethoden. Mit der „truncated product
method” wird dabei ein neuer Ansatz dargestellt. Eine Kombinationsmethode wird u.
a. für ein adaptives Design benötigt. Die Möglichkeiten, die ein adaptives Studiendesign
bietet, sind Thema des letzten Abschnitts.
Abstract
This paper presents some novel biometrical methods. The first topic is the comparison
of two groups. Beside the Wilcoxon rang sum test two recent tests are presented, i.
e. the Baumgartner-Weiß-Schindler test and the Brunner-Munzel test. In particular,
the situation of heterogeneous variances is discussed. The second part discusses combination
methods and, especially, the recently introduced truncated product method. A combination
test is necessary within an adaptive design. Thus, the last section presents possibilities
an adaptive study design offers.
Schlüsselwörter
Statistik - Zweistichprobentests - nichtparametrische Methoden - Kombinationsmethoden
- adaptive Designs
Key words
Statistics - two-sample tests - nonparametric methods - combination methods - adaptive
designs
Literatur
- 1
Büning H.
Robust analysis of variance.
Journal of Applied Statistics.
1997;
24
319-332
- 2
Nanna M J, Sawilowsky S S.
Analysis of Likert scale data in disability and medical rehabilitation research.
Psychological Methods.
1998;
3
55-67
- 3 Büning H, Trenkler G. Nichtparametrische statistische Methoden. 2. Auflage. Berlin;
De Gruyter 1994
- 4 Zimmerman D W, Zumbo B D. The relative power of parametric and nonparametric statistical
methods. Keren G, Lewis C A handbook for data analysis in the behavioral sciences:
methodological issues Hillsdale/NJ; Lawrence Erlbaum Associates 1993: 481-517
- 5
Sawilowsky S S, Blair R C.
A more realistic look at the robustness and type II error properties of the t test
to departures from population normality.
Psychological Bulletin.
1992;
111
352-360
- 6
Baumgartner W, Weiß P, Schindler H.
A nonparametric test for the general two-sample problem.
Biometrics.
1998;
54
1129-1135
- 7
Neuhäuser M.
Exact tests based on the Baumgartner-Weiß-Schindler statistic - a survey. Statistical
Papers (im Druck).
- 8
Ogenstad S.
The use of generalized tests in medical research.
Journal of Biopharmaceutical Statistics.
1998;
8
497-508
- 9
Brunner E, Munzel U.
The nonparametric Behrens-Fisher problem: asymptotic theory and a small sample approximation.
Biometrical Journal.
2000;
42
17-25
- 10
Neuhäuser M, Poulin R.
Comparing parasite numbers between samples of hosts.
Journal of Parasitology (im Druck).
- 11 Brunner E, Munzel U. Nichtparametrische Datenanalyse. Berlin; Springer 2002
- 12 Neuhäuser M, Lam F C. Nonparametric approaches to detecting differentially expressed
genes in replicated microarray experiments. Chen YPP Bioinformatics 2004: Proceedings
of the 2nd Asia-Pacific Bioinformatics Conference. Conferences in Research and Practice
in Information Technology Sydney; Australian Computer Society 2004 29: 139-143
- 13
Wallis W A.
Compounding probabilities from independent significance tests.
Econometrica.
1942;
10
229-248
- 14
Zaykin D V, Zhivotovsky L A, Westfall P H. et al .
Truncated product method for combining P-values.
Genetic Epidemiology.
2002;
22
170-185
- 15
Neuhäuser M.
Tests for genetic differentiation.
Biometrical Journal.
2003;
45
974-984
- 16
Dudbridge F, Koeleman B PC.
Rank truncated product of P-values, with application to genomewide association scans.
Genetic Epidemiology.
2003;
25
360-366
- 17
Bauer P, Köhne K.
Evaluation of experiments with adaptive interim analyses.
Biometrics.
1994;
50
1029-1041
- 18
Neuhäuser M.
Adaptive Designs und die „truncated product method”.
Informatik, Biometrie und Epidemiologie in Medizin und Biologie.
2003;
34
314-315
- 19
Neuhäuser M.
An adaptive interim analysis - a useful tool for ecological studies.
Basic and Applied Ecology.
2001;
2
203-207
- 20
Bauer P, Brannath W, Posch M.
Flexible two-stage designs: an overview.
Methods of Information in Medicine.
2001;
40
117-121
Dr. rer. nat. Markus Neuhäuser
Institut für Medizinische Informatik, Biometrie und Epidemiologie, Universitätsklinikum
Essen
Hufelandstraße 55
45122 Essen
Telefon: 02 01/7 23-44 62
Fax: 02 01/7 23-59 33
eMail: markus.neuhaeuser@medizin.uni-essen.de