Optimal Experimental Design

A wide class of problems of optimal experimental design is studied. Recent directions of interest of the cluster include:

• Investigation of multiplicative algorithms of construction of D-optimal designs
• Optimal designs in the presence of correlated observations with long-range dependence
• Optimal design in mixed and random effect models
• Optimal allocation of doses in the dose-finding studies
• Design in multi-centre clinical trials
• Adaptive design of experiments for seeking the optimal conditions
• Optimal design of computer experiments
• Probabilistic existence theorems in group testing

Cardiff Investigators

• Prof Anatoly Zhigljavsky
• Dr Andrey Pepelyshev
• Prof Nikolai Leonenko

Collaborators

• Prof Henry Wynn
• Prof Valerii Fedorov
• Prof Viatcheslav Melas
• Prof Anthony Atkinson
• Dr Luc Pronzato
• Dr Ben Torsney
• Prof Michail Malyutov
• Dr Radoslav Harman

Selected publications

• Atkinson A.C., Bogacka B., Zhigljavsky A.A., eds. (2001) Optimum Design - 2000, Kluwer Academic Publishers, x+306 pp.
• Muller W., Wynn H.P., Zhigljavsky A.A., eds. (1993) Model Oriented Data Analysis, Physica-Verlag, Berlin, xiii+287 pp.
• Ermakov S.M., Zhigljavsky A.A. (1987) Mathematical Theory of Optimal Experiment. Nauka, Moscow, 320 pp. (in Russian).


• Dette H, Pepelyshev A, Zhigljavsky A. (2014) Design for linear regression models with correlated errors. Handbook of Design and Analysis of Experiments
• Dette H., Pepelyshev A., Zhigljavsky A. (2013). Optimal design for linear models with correlated observations. Annals of Statistics, 41, 143-176. link
• Dette H., Holland-Letz T., Pepelyshev A. (2011) A geometric characterization of optimal designs for regression models with correlated observations. JRSS B, 73, 239–252. link
• A. Zhigljavsky, H. Dette, A. Pepelyshev (2010) A new approach to optimal design for linear models with correlated observations. J. Amer. Stat. Ass. 105, 1093-1103.link
• H. Dette, N. Leonenko, A. Pepelyshev, A. Zhigljavsky (2009) Asymptotic optimal designs under long-range dependence error structure. Bernoulli 15, 1036-1056.link
• H. Dette, A. Pepelyshev, A. Zhigljavsky (2008) Improving updating rules in multiplicative algorithms for computing D-optimal designs. Com. Stat. and Data Analysis. 53, 312-320.link
• Zhigljavsky A.A. (2003) Probabilistic existence theorems in group testing, Journal of Statistical Planning and Inference, vol. 115, No. 1, 1 - 43.
• Zhigljavsky A.A., Zabalkanskaya L. (1996) Existence theorems for some group testing strategies, Journal of Statistical Planning and Inference, 55, No 2, 151-173.