1-Cox DR e Reid N, 2000, The theory of the design of the experiment, Chapman & Hall- (chap. n. 1 e n.2)
2-Montgomery DC, 1991, Design and analysis of experiment, Wiley- (chapters n. 6,8,9,11). Chapter 11 until the II order, including blocks.
3-Khuri I e Cornell JA, 1987, Response surfaces: design and analyses, Marcel Dekker- (chap. n. 1, 5, 10).
4-Berni R.,2014 working paper n.10;http://local.disia.unifi.it/wp_disia/2014/wp_disia_2014_10.pdf
5-Atkinson A.C., Donev A.N., 1992, Optimum experimental design, Oxford Statistical Science
Series, Clarendon Press- (chapters n.1, 5, 9, 20).
Regarding Choice-experiments, lectur notes and papers will be supplied by the teacher.
Learning Objectives
Deep knowledge of the design of experiments: starting from the classical experimental design up to the recent developments, by considering theory and applications in the technological and industrial fields.
Prerequisites
Statistical inference; mathematical analysis; linear algebra
Teaching Methods
Theory; laboratory;
and specific task for each student (on real data)
Further information
1- For the student who did not attend a course of design of experiments in his/her previous degree, he/she must read the teacher's notes (dispense.pdf) available on the teacher's web-site.
Type of Assessment
Oral examination.
Questions will be related to the main arguments of the course, as described in the diploma supplement. Particular attention is payed to the critical and constructive student's abilities
Course program
Starting from the local approximation in Taylor's series, the fundamental theory of response surface methodology is introduced, by considering the I and II order. In particular: polynomial models, designs and properties, the moment array. A specific attention is devoted to optimization methods, I and II order. The experimental designs are: fractional factorials (at 2, 3 and mixed levels), Central Composite design.
The second part is related to optimal design by considering the general issues and the D, G, T criteria.
The third part is related to the split-plot design, following the last features and developments of this plan. In this context, the concept of random effect is introduced through the mixed Response Surface models.
Finally, specific attention is payed to the choice-experiments, by considering the optimal designs and the specific characterisstics of design pf experiments when applied to the consumer's behaviour evaluation.
FINAL REMARK:
during the initial part (8 hours) of the course, specific attention is devoted to students who attended the Statistical Quality Control examination in the first level Statistical Degree. For these students, extended topics will be supplied.