Ajit Tamhane

Ajit C. Tamhane is a professor in the Department of Industrial Engineering and Management Sciences (IEMS) at Northwestern University and also holds a courtesy appointment in the Department of Statistics.

Tamhane has published over 100 research articles in refereed journals and has authored four books and co-edited two volumes of collected research papers. His research primarily focuses on multiple testing in clinical trials. He has also worked extensively in other areas of statistics including design of experiments, ranking and selection procedures, chemometrics, clustering methods and statistical inference.

Tamhane is a fellow of the American Statistical Association, the Institute of Mathematical Statistics, the American Association for Advancement of Science and an elected member of the International Statistical Institute.

Education Tamhane studied at the Indian Institute of Technology Bombay, and received his B.Tech. in Mechanical Engineering in 1968. He moved to the United States in 1970, earning his Ph.D. in Operations Research and Statistics from Cornell University in 1975 under the supervision of Robert E. Bechhofer. Statistical Analysis of Designed Experiments, Predictive Analytics: Parametric Models for Regression and Classification Using R, and Multiple Comparison Procedures. He has also edited two volumes of collected papers: Design of Experiments: Ranking and Selection (with Thomas Santner) published by Marcel Dekker (1984) and Multiple Testing Problems in Pharmaceutical Statistics (with Alex Dmitrienko and Frank Bretz) published by Chapman & Hall (2010).

Multiple comparisons Tamhane provided and explored several test procedures for the identification of the minimum effective and maximum safe doses of a drug (MINED and MAXSD). He also studied the usage of adaptive extensions of a two-stage group sequential procedure (GSP) in terms of testing primary and secondary endpoints, and discussed different ways to modify the boundaries of the original group sequential procedure to control the familywise error rate, and also provided power comparisons between competing procedures along with clinical trial examples. In his paper published in 2011, he defined classes of parallel gatekeeping procedures. Results of his study indicated an improvement in power of multistage gatekeeping procedures by the usage of α-exhaustive tests for the component procedures. Eric Peritz reviewed Tamhane's book entitled, Multiple Comparison Procedures, as "a comprehensive monograph" in which "the control of familywise error rates is given the lion's share in the book."

Design of experiments for selection and multiple testing problems Tamhane's early work, emanating from his Doctoral Dissertation, was on two-stage and multi-stage screening type procedures for selecting the best treatment. He studied the design of such procedures, focusing on the sample size requirements. For the problem of testing multiple treatments with a common control, he generalized the classical balanced incomplete block (BIB) designs to what are called balanced treatment incomplete block (BTIB) designs.

Chemical engineering applications, quality control and data mining In his study regarding chemical engineering applications, Tamhane proposed a novel nonparametric regression method for high-dimensional data, nonlinear partial least squares (NLPLS), and implemented it with feedforward neural networks. He further determined the performances of NLPLS, projection pursuit, and neural networks in the context of response variable predictions and robustness to starting values. He also conducted multiple studies regarding the detection of gross errors in process data in chemical process networks.

Awards and honors 1985 - Youden Award for the best expository paper published in Technometrics 1991 - Fellow, American Statistical Association 2010 - Fellow, Institute of Mathematical Statistics