Politecnico di Torino
Politecnico di Torino
   
Login  
it
Politecnico di Torino
Academic Year 2015/16
02PVFMQ, 02PVFLX, 02PVFLZ, 02PVFNZ, 02PVFOA, 02PVFPC
Business analytics
1st degree and Bachelor-level of the Bologna process in Mathematics For Engineering - Torino
1st degree and Bachelor-level of the Bologna process in Electrical Engineering - Torino
1st degree and Bachelor-level of the Bologna process in Aerospace Engineering - Torino
Espandi...
Teacher Status SSD Les Ex Lab Tut Years teaching
Brandimarte Paolo ORARIO RICEVIMENTO PO MAT/09 40 20 0 0 6
SSD CFU Activities Area context
SECS-S/01 6 D - A scelta dello studente A scelta dello studente
Subject fundamentals
Recent years have been characterized by the explosive growth in the application of high-level, commercial statistical tools (e.g., SPSS and SAS) in various application domains, such as finance, marketing, and retail management. This course aims at providing the student possessing a strong mathematical background with the practical skills in order to fruitfully apply this wide range of techniques to the three business analytics levels (descriptive, predictive, and prescriptive), integrating probability, statistics and optimization methods.

The course objectives are:
- To illustrate the application of methods learned in basic statistics through the discussion of concrete business cases.
- To introduce fundamental concepts for decision making under uncertainty.
- To introduce essential multivariate analysis methods, with emphasis on data reduction.
- To present some relevant examples of applications to management (revenue/yield management; risk measurement and management; market analysis and forecasting; customer relationship management).
Expected learning outcomes
Knowledge:
- Multivariate statistical methods.
- Methods and models for decision making under uncertainty.
- Knowledge of standard problem formulations and solution methods for managerially relevant application domains (service and product pricing, yield management, retail management, risk management, market segmentation).
Skills:
- Ability to analyze real datasets.
- Ability to apply statistical and decision models in real management contexts.
- Use of R.
Prerequisites / Assumed knowledge
This course has a strong orientation toward quantitative analysis and assumes a deep familiarity with the concepts provided by basic course in calculus and linear algebra (e.g: Taylor expansions; convex functions; quadratic forms; eigenvalues and eigenvectors; matrix diagonalization and orthogonal matrices; constrained optimization; numerical integration). It is absolutely necessary to have followed a basic course on probability and statistics (discrete and continuous random variables; multivariate distributions; inferential statistics: point and interval parameter estimation and hypothesis testing; linear regression models). A previous exposure to operations research is very useful, but not strictly required.
Furthermore, given the extensive use of R, the familiarity with scientific computing environments (like MATLAB) is also essential, as well as some basic programming skills.
Contents
Using R for statistical applications (10 hours).
Descriptive statistics. Random number and variate generation. Parameter estimation. Hypothesis testing. Correlation analysis. Simple linear regression.

Decision making under uncertainty (20 hours).
Risk and uncertainty. Decision trees, investment decisions, and real options. Stochastic linear programming and robust optimization. Monte Carlo simulation. Utility theory and risk measures.

Statistical multivariate analysis methods (20 hours).
Multiple linear regression. Logistic regression. Principal Component Analysis. Factor analysis. Cluster analysis. Regression trees. Multidimensional scaling and correspondence analysis.

Discussion of business case involving the application of statistical methods (10 hours).
Possible examples are: Colonial Broadcasting Company (linear regression); Paper and more (demand forecasting and inventory management under uncertainty); Pilgrim Bank (customer profitability and retention analysis by linear and logistic regression).
Delivery modes
A significant role is played by the discussion of business cases, mostly published by Harvard Business School Publishing. Furthermore, we will make extensive use of the statistical software R, which is freely available for download. It is useful (but not required) to install R on a laptop for classroom use.
Texts, readings, handouts and other learning resources
Course slides will be available.
There is no single textbook but most slides are taken from: P. Brandimarte. Quantitative methods: An introduction for business management. Wiley, 2011.
The list of business cases to be discussed will be posted online.
Assessment and grading criteria
Written exam (90 minutes), including numerical problems, questions about theory, and simple proofs, possibly connected with the business cases that are discussed in class.

Programma definitivo per l'A.A.2015/16
Back



© Politecnico di Torino
Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY
WCAG 2.0 (Level AA)
Contatti