Michael G. Akritas

Professor of Statistics
Ph.D., University of Wisconsin-Madison, 1978

Summary of research interests

The research efforts of Dr. Akritas are focused on three areas.

The first area involves nonparametric methods for factorial designs. The principal development here is the introduction of new concepts for interaction effects, main effects, and simple effects, which are called nonparametric effects. The main advantage of these effects over the traditional parametric effects is that the associated null hypotheses are invariant under monotone transformations. This allows straightforward analysis and interpretation of data especially when the normal assumption that underlies the traditional theory for linear models is not met.

The second area is on regression and other methods with incomplete data. The incomplete data is considered in Dr. Akritas' research arise due to censoring and/or truncation. Such data are common in medical research, sociology, astronomy, and ecology.

The final research area involves the measurement error model introduced by Dr. Akritas for data that arise in astronomy. In the context of this measurement error model, there are a number of interesting problems such as deconvolution, goodness-of-fit, and variance function estimation in regression.

Finally, Dr. Akritas is the director of the National Statistical Consulting Center for Astronomy (SCCA) which is funded by NASA.

Representative publications

Akritas, M. G. (1997). On the Use of Nonparametric Regression Techniques for Fitting Parametric Regression Models. Biometrics>

Akritas, M. G. and Bershady, M. A. (1997). Linear Regression for Astronomical Data with Measurement Errors and Intrinsic Scatter. The Astrophysical Journal.

Akritas, M. G. and Siebert, J. (1996). Testing for Partial Association Using Kendall's T with Censored Astronomical Data. Monthly Notices of the Royal Astronomical Society.

Akritas, M. G., Murphy, S. A., and LaValley, M. P. (1995). The Theil-Sen Estimator with Doubly Censored Data and Applications to Astronomy. J. Amer. Statist. Assoc. 90, 170 -177.

Akritas, M. G. (1994). Nearest Neighbor Estimation of a Bivariate Distribution Under Random Censoring. Ann. Statist. 22, 1299 - 1327.

Akritas, M. G. and Arnold, S. F. (1994). Fully Nonparametric Hypotheses for Factorial Designs I: Multivariate Repeated Measures Designs. J. Amer. Statist. Assoc. 89, 336 - 343.