Biostatistics; Models, Statistical
My main research focus is on biostatistics and the planning, design and analysis of clinical trials.
Especially my fields of interest include adaptive designs, where design modifications may be performed after an interim analysis, and multiple testing procedures in the context of clinical trials and bioinformatics.
Techniques, methods & infrastructure
As a medical statistician I am involved in numerous collaborative research projects where I contribute to the planning, analysis and interpretation of clinical trials. Working on applied clinical trials is often leading to new research questions in statistical methodology. E.g. multiple testing refers to the testing of more than one statistical hypothesis at the same time. The problem of multiple testing arises in several areas of medical research. Examples are subgroup analyses, analyses of large numbers of variables as in gene expression or proteomic studies, or hypotheses tests for adaptive trials, where design modifications may be performed after an interim analysis.
My research includes the investigation of potential Type 1 Error inflation when using conventional statistical analysis strategies ignoring possible multiple testing issues or reflecting existing methodology that is used under a new context. A further important issue is to construct new design methodology controlling the Type 1 Error if existing methodology is not appropriate.
Complete list of my publications
- Type 1 Error Rate Inflation in Multi-armed Clinical Trial (2012)
Source of Funding: FWF (Austrian Science Fund), Erwin-Schrödinger Fellowship
- Graf AC., Wassmer G., Friede T., Gera RG., Posch M. (2019): Robustness of testing procedures for confirmatory subpopulaton analyses based on a continuous biomarker. Stat Methods Med Res, 28 (6): 1879-1892
- Graf AC., Gutjahr G., Brannath W. (2016): Precision of Matimum Likelihood Estimation in Adaptive Designs, Stat in Med, 25 (1): 922 - 941.
- Graf AC., Posch M., Koenig F. (2015): Adaptive designs for subpopulation analysis optimizing utility functions, Biom J, 57 (1), 76 - 89.
- Graf AC., Bauer P., Glimm E., Koenig F. (2014): Maximum Type 1 Error Rate Inflation in multi-armed clinical trials with interim sample size modifications, Biom J, 56 (4) 614 ,
- Graf AC., Bauer P. (2011): Maximum inflation of the type 1 error rate when sample size and allocation rate are adapted in a pre-planned interim look. Stat in Med, 30 (14): 1637-1647.