DATA SCIENCE FOR MARKETING
cod. 1007332

Academic year 2020/21
2° year of course - First semester
Professor
Andrea CERIOLI
Academic discipline
Statistica (SECS-S/01)
Field
Statistico-matematico
Type of training activity
Characterising
68 hours
of face-to-face activities
9 credits
hub: PARMA
course unit
in ITALIAN

Learning objectives

The course gives knowledge on statistical techniques for Marketing applications and for the analysis of consumer behavior. These techniques include:
multiple linear regression; logistic regression; classification trees; nonhierarchical
cluster analysis.

The aim of the course is threefold:
1. To provide both a theoretical and a practical understanding of the key methods of model building, classification and prediction.
2. To provide a Marketing-driven context for these methods.
3. Using real data and case studies, and a learning-by-doing approach, to illustrate
the application and the interpretation of these methods.

Computational aspects of the methods are also addressed through the use of MS Excel and IBM SPSS.

Prerequisites

Knowledge of basic statistical methods, as given in undergraduate
programs in Economics and Management. Knowledge of the content of the course "Statistical Methods for Management".

Course unit content

The aim of this course is to explain the main statistical techniques that
are useful for Marketing applications to large data bases, with emphasis on consumer behaviour. Specifically, the course covers:
a) multiple linear regression and its applications in Marketing;
b) logistic regression for the prediction of consumer behaviour;
c) classification trees for the prediction of consumer behaviour and for consumer segmentation;
d) cluster analysis for consumer segmentation.
The course covers both the tatistical theory behind these techniques and their applications, by means of practical work through MS Excel and IBM SPSS.

Full programme

- - -

Bibliography

M. Riani, F. Laurini, G. Morelli: Strumenti statistici e informatici per applicazioni aziendali. Pitagora Editrice, Bologna, 2013, from Chapter 4 onwards (with the exception of Sections 4.6, 4.12, 5.3, 5.4 and Appendix A). Also available through the web site: https://editrice.pitagoragroup.it/

A. Cerioli, F. Laurini: Il modello di regressione logistica per le applicazioni aziendali, Uni.Nova, Parma, 2019 (all with the exception of Section 5.3.1 and Appendix A). Also available through the web site: https://www.uninova.it/

S. Zani e A. Cerioli: Analisi dei dati e data mining per le decisioni
aziendali, Giuffrè, Milano, Chapter IX (Sections 1 – 2 – 11 – 12), Chapter XI (with the exception of Sections 4, 5.4 and 6). Also available through the web site: https://shop.giuffre.it/

Course slides.

Teaching methods

Distance learning through online and offline lectures (58 hours), web forums, and problem-oriented meetings (10 hours).

Teaching materials (course slides and data for replicating analyses) are provided through Elly.

Further material (data for individual practical work and research readings) is also provided through Elly, as well as updated details on the course timetable.

Course slides are also available at the Printing Office.

Assessment methods and criteria

Written exam of 60 minutes, at which the student can bring the textbooks and a calculator.

In the current distance learning scenario, the exam will be performed trhough the Elly tool named COMPITO and under the available surveillance tools (e.g. Teams or Zoom).

Knowledge and understanding are assessed by
methodological questions, marked 3 grade points each. The ability of
applying knowledge and understanding are assessed by questions on
the interpretation of results, marked 3 grade points each. Learning skills
are assessed by questions on the conclusions to be drawn from an
analysis, marked 3 grade points each.

Details on examination procedures are provided to the class and made available through Elly at the start of the course.

Other information

- - -