Another area of interest is statistical computing methodology and algorithms and the human interface with statistical programs. He has worked with students in developing the algorithms for balanced, orthogonal analysis of variance and more general linear model routines that provide hypothesis tests for messy, unbalanced data often encountered in observational studies. These routines appear in the widely used MINITAB statistical package.
Recent research with graduate students developed sequential design algorithms for the linear logistic regression and other nonlinear models. These models are used to represent the probability of an event as a function of a dose or stimulus. The sequential procedure utilizes the experimental response at preceding stages to optimally set the dose at subsequent stages.
Representative Publications:
Rosenberger, J. L., and M. Gasko. 1982. Comparing location estimators: trimmed means, medians, and trimean. In Understanding Robust and Exploratory Data Analysis, edited by D. C. Hoaglin, F. Mosteller, and J. W. Tukey. J. Wiley & Sons.
Hill, R. R., and J. L. Rosenberger. 1985. Methods for combining data from germ-plasm evaluation trials. Crop Science 25:467-470.
Rosenberger, J. L. 1989. Design of experiments for evaluating frozen sperm. Journal of Andrology 11:89-96.
Vasilatos-Younken, R., K. S. Gray, W. L. Bacon, K. E. Nestor, D. W. Long, and J. L. Rosenberger. 1990. Ontogeny of growth hormone (GH) binding in the domestic turkey: Evidence of sexual dimorphism and developmental changes in relationship to plasma GH. Journal of Endocrinology 126:131-139.
Pukelsheim, F. and J. L. Rosenberger. 1993. Experimental designs for model discrimination. Journal of the American Statistical Association 88:642-649.