[Forum SIS] Seminario Prof. Philip J Brown (University of Kent), Università Ca' Foscari di Venezia, 24 maggio 2016 ore 11.00

[Debora Slanzi]

***********************************************************
AVVISO DI SEMINARIO
***********************************************************

Data: 24/05/2016, Ora: 11:00
Luogo: Dipartimento di Scienze Ambientali, Informatica e Statistica, Università Ca’ Foscari di Venezia 
Sala Conferenze, Campus Scientifico di via Torino (Mestre) 

Presenta:
Philip J Brown, University of Kent (UK)

Organizzatore:
Debora Slanzi

Titolo:
Level-Screening designs in drug discovery

Abstract:
We look at applying classical design ideas to situations where there is a very low information content, as arises in areas like microarrays and more specially drug screening. Surprisingly some of these ideas do not give good designs and the best designs are very unbalanced. We find optimal designs when there are few factors but many levels to each factor and relatively few observations. This is typical in drug screening when experimentation proceeds sequentially and clues to good future compounds are sought. We study in detail two designs, the Dumbbell and Sawtooth and also investigate their robustness to missing data.


Bio Sketch
Philip J Brown is currently Emeritus Professor of Medical Statistics at University of Kent. He has served as Lecturer in Statistics, Imperial College London (1971-85), Professor of Mathematical Statistics University of Liverpool, 1985-95, Pfizer Professor of Medical Statistics, University of Kent 1995-2013. He also spent time abroad: Berkeley, California 1975-6, Helsinki, Finland in 1985, CSIRO Melbourne, Australia 2003-4. He is a fellow of the American Statistical Association and the the Institute of Mathematical Statistics. In 2003 he was a co-recipient of the Mitchell Prize in Bayesian Statistics awarded by the American Statistical Association. He has collaborated with researchers in USA, Canada, Venezuela,  Australia, Finland, Sweden and Italy. He is author of ‘Measurement, Regression and Calibration’ (Oxford University Press). He has published papers  in a wide range of applied areas ranging from Election Night forecasting to medical diagnosis with core methodologies in Bayesian multivariate analysis focussing especially on feature selection in complex high-dimensional settings.




Tutti gli interessati sono invitati a partecipare.
Cordiali saluti,
Debora Slanzi

_____________________________________
Debora Slanzi, Ph.D.
Department of Environmental Sciences, 
Informatics and Statistics
University Ca'Foscari of Venice
Phone: +39 041 234 7594
Fax: +39 041 234 8584
Email: debora.slanzi a unive.it <mailto:debora.slanzi a unive.it>
Web: www.unive.it/persone/debora.slanzi <http://www.unive.it/persone/debora.slanzi> 

***A mathematician, like a painter or a poet, is a maker of patterns (G.H. Hardy, 1941)***

-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20160520/2f0ccc22/attachment-0001.html>