Mathematical Methods and Probability

PROFF. CARLO A. DE BERNARDI, GIULIA GIANTESIO, ENRICO MIGLIERINA

COURSE CONTENT

The course is divided into four parts.

Probability 1. (Giulia Giantesio – 10 hours). Events and probability. Random
variables: discrete and continuous. Expectation, variance, and conditional
expectation. Main univariate probability distributions. The change-of-variable
formula.

Linear Algebra. (Enrico Miglierina – 20 hours) Brief review of definitions and basic
operations of vectors matrices. Trace, determinant and inverse of square matrices.
Eigenvalues and Eigenvectors. Definite, semi-definite and indefinite matrices
(including Cholesky Decomposition). Quadratic Forms and the associated matrix.

Calculus. (Carlo A. De Bernardi – 20 hours).

Improper integrals. 𝑅 n as normed space. Differential calculus for functions of several
variables. Taylor approximation. Unconstrained optimization. Constrained
optimization (Lagrange multipliers). Double integral.

Probability 2. (Giulia Giantesio – 10 hours) Random vectors and multivariate

distributions. Expectation vector, variance-covariance matrix, correlation matrix.
Multivariate normal and Multinomial distributions. Convergence of random
variables, law of large numbers, central limit theorem.

ASSESMENT METHOD-EXAM RULES

The exam is written and the exam dates are planned according to the university’s
calendar:
November 21, 2022,
December 12, 2022,
March 13, 2023
June 6, 2023
September 4, 2023
Each exam consists of two parts:
 Part 1, containing 12 questions of the following types: multiple choice,
true/false, numerical answer,…; this part is 1 hour long;
  Part 2, containing open questions requiring an elaborated solution. The total
amount of points obtainable in this part is 32; this part is 1 hour and 40
minutes long.
The exam is passed with at least 8 correct answers in Part 1 and at least 18 points in
Part 2. The grade obtained in the exam is the amount of points obtained in Part 2,
i.e., the grade does not depend on the number of correct answers given in Part 1.
Those who obtain more than 30 points receive as grade 30 e lode.
The commission reserve the right, in certain and particular situations, to:
 apply in a less restrictive way the rules described above;
 require an oral integration to complete the exam.
–The exams will be performed (in the presence or online) depending on the health
situation related to the Covid-19 pandemic with a communication before the exam.
In the case the exam will take place online, all the details will be given before the
exams.
To take part in the exam the student should enroll to the exam by I-Catt online
system.

READING LIST
 Marc Peter Deisenroth, A. Aldo Faisal, and Cheng Soon Ong -Mathematics for Machine
Learning, Cambridge University Press, 2020.
 L. Peccati, S. Salsa, and A. Squellati, Mathematics for Economics and Business,
Bocconi University Press, Milan, 2016.

TEACHING METHOD
Lectures in presence, exercise groups, tutoring hours.

NOTES AND PREREQUISITES

Prerequisites:
The student is assumed to know differential calculus in R and some basic notions of linear
algebra (vectors and matrices, linear systems).

Announcements and additional material will be available on Blackboard

Some tutoring meetings, announced on Blackboard, will be organized in the
second part of the course.

Office hours: by appointment by contacting the teachers via e-mail:

Carlo Alberto De Bernardi: carloalberto.debernardi@unicatt.it
Giulia Giantesio: giulia.giantesio@unicatt.it
Enrico Miglierina: enrico.miglierina@unicatt.it