84611 - Social Statistics

Learning outcomes

At the end of the course, students will have acquired knowledge of and be able to apply statistical methods for market analysis. In particular, they will be able to: -design and implement surveys to elicit consumers' preferences for alternative products; - estimate, evaluate, and interpret discrete choice models and market response models.

Course contents

The course will cover the following topics:

Theoretical foundations of discrete choice models: behavioral theories and model derivation.

Data for choice analysis: revealed and stated preferences and data sources.

Design and implementation of discrete choice experiments: experimental design and questionnaire development.

Discrete choice models: multinomial logit, mixed logit and latent class models. For each model, the course covers the theoretical foundations, model specification, estimation, interpretation of results, and practical applications.

Managerial applications: estimation of elasticities, market share forecasting, and counterfactual scenario simulations, with particular emphasis on the effects of price changes.

Prerequisites: Basic knowledge of regression models and binary logit models, familiarity with the R programming language, and a basic understanding of marketing principles.

Readings/Bibliography

Teaching materials will be made available on the Virtuale platform.

The main reference texts are:

Bassi, F., & Ingrassia, S. (2022). Statistica per analisi di mercato. Metodi e strumenti. Pearson, Milan.

Train, K. E. (2009). Discrete Choice Methods with Simulation. Cambridge University Press.

Shang, L., & Chandra, Y. (2023). Discrete Choice Experiments Using R. Springer Singapore.

Teaching methods

The course combines lectures, instructor-guided group work, and computer laboratory sessions devoted to the development of case studies. Applications are based on real-world data and implemented in R.

Given the nature of the course activities and teaching methods, all students are required to complete Modules 1 and 2 of the mandatory e-learning training on health and safety in study environments before attending the course.

For further information, see the mandatory health and safety training courses.

Assessment methods

For attending students, assessment is based on the completion of a group project designed to apply the knowledge and skills acquired during the course.

The group project (3–4 students per group) consists of developing a business case on the elicitation and analysis of consumer preferences. Students will identify a research topic, design a questionnaire including a discrete choice experiment, collect data through interviews, analyse the data using discrete choice models, and discuss the managerial implications of the results.

The project will be presented orally, and all group members are expected to present part of the work. During the presentation, the instructor may ask individual questions about both the project and any topic covered in the course.

The final individual grade is based on the overall quality of the project, the quality of the individual presentation, and the answers to the individual questions. Assessment follows these criteria: 28–30 cum laude: accurate, complete, and well-structured work demonstrating full mastery of the course contents, strong analytical and interpretative skills, and the ability to apply the acquired knowledge to real-world cases; 24–27: generally accurate work with some minor inaccuracies, showing good knowledge of the course contents and adequate analytical and interpretative skills; 18–23: work containing methodological or analytical weaknesses but meeting the minimum requirements for a passing grade.

For assessment purposes, limited, declared, and non-substantive use of Artificial Intelligence (AI) tools is permitted for support activities (e.g. code debugging and the preparation of tables, figures, and infographics for presentation slides). The use of AI to complete substantial parts of the assessed work is not permitted.

----------------------------------------------------------------------

Non-attending students are assessed through a written examination.

The written exam consists of open-ended questions and applied exercises. Each question is worth between 2 and 5 points, depending on its type and level of difficulty. The final score is the sum of the points obtained on each question. The maximum score is 31, corresponding to 30 cum laude. The minimum passing grade is 18/30. Indicative grading is as follows: 30–31: excellent; 27–29: very good; 24–26: good; 21–23: satisfactory; 18–20: sufficient.

The exam includes two types of questions: theoretical and applied. Theoretical questions assess knowledge, accuracy, completeness, correct use of terminology, and clarity of exposition. They may require either short or extended answers. Applied questions assess the ability to interpret model outputs correctly, relate the results to the underlying theory and assumptions, discuss their implications and limitations, and present answers clearly. The wording of each question explicitly states the required task (e.g. define, interpret) to ensure transparency and consistency.

The written examination lasts 100 minutes. During the exam, the use of textbooks, notes, electronic devices, or any other supporting materials is not permitted. The use of Artificial Intelligence (AI) is strictly prohibited. Any use of AI constitutes a violation of academic integrity.

Teaching tools

Lecture slides, recommended readings, datasets, and R code for the practical sessions will be made available on the Virtuale platform.

Office hours

See the website of Beatrice Biondi