LABORATORY OF STATISTICAL DECISIONS

Course objectives

Learning goals Develop analytical and computational skills to solve statistical decision problems. Knowledge and understanding At the end of the course the students have the ability to understand and solve simple and more advanced exercises of Statistical decision theory. Applying knowledge and understanding Students are required to apply theoretical and computational skills (using the software R) to solve inferential problems formalized as decision problems. Making judgements One of the main goals of practical activities is to develop the ability of comparing and choosing alternative methods, i.e. to refine judgement skills. Communication skills Students acquire the ability of presenting written reports of their practical laboratories. Learning skills The students acquire a series of skills useful for future academic and professional activities.

Channel 1
FULVIO DE SANTIS Lecturers' profile

Program - Frequency - Exams

Course program
Use of software R for statistical decision problems. Part 1 (1/3 of the course) Bayesian inference in R (statistical software). Prior and posterior distributions; point and interval estimation; test; predictive analysis; normal approximations; simulation (Monte Carlo). Part 2 (2/3 of the course) Analysis of decisions and of statistical decisions with R. Loss functions; risk function; posterior expected loss; optimal decisions; simulation.
Prerequisites
Probability (randoma variables, distributions, limit theorems), Statistical inference (frequentist theories of estimation and testing) and Decision theory (frequentist and Bayesian statistical decisions).
Books
Provided by the instructor and available on the Moodle web page of the class "Statistical decision theory": https://elearning.uniroma1.it/course/view.php?id=19894 (pwd: decisioni2025)
Teaching mode
Lectures and in-class activity (with software R). lecture will be given in class unless forbidden by sanitary restrictions
Frequency
Required (at least 9/12 labs). Attendance is registered in class.
Exam mode
One written practical exam using R at the end of the semester. Duration: 60 minutes. Students must anwer at least 50% of the questions plus one. Types of problems: numerical problems, data analysis, simulations.
Bibliography
De Santis F. (2025). Basic Bayes (available on Moodle). Piccinato L. (2009). Metodi per le decisioni statistiche. Springer
Lesson mode
Lectures and in-class activity with software R.
FULVIO DE SANTIS Lecturers' profile

Program - Frequency - Exams

Course program
Use of software R for statistical decision problems. Part 1 (1/3 of the course) Bayesian inference in R (statistical software). Prior and posterior distributions; point and interval estimation; test; predictive analysis; normal approximations; simulation (Monte Carlo). Part 2 (2/3 of the course) Analysis of decisions and of statistical decisions with R. Loss functions; risk function; posterior expected loss; optimal decisions; simulation.
Prerequisites
Probability (randoma variables, distributions, limit theorems), Statistical inference (frequentist theories of estimation and testing) and Decision theory (frequentist and Bayesian statistical decisions).
Books
Provided by the instructor and available on the Moodle web page of the class "Statistical decision theory": https://elearning.uniroma1.it/course/view.php?id=19894 (pwd: decisioni2025)
Teaching mode
Lectures and in-class activity (with software R). lecture will be given in class unless forbidden by sanitary restrictions
Frequency
Required (at least 9/12 labs). Attendance is registered in class.
Exam mode
One written practical exam using R at the end of the semester. Duration: 60 minutes. Students must anwer at least 50% of the questions plus one. Types of problems: numerical problems, data analysis, simulations.
Bibliography
De Santis F. (2025). Basic Bayes (available on Moodle). Piccinato L. (2009). Metodi per le decisioni statistiche. Springer
Lesson mode
Lectures and in-class activity with software R.
  • Lesson codeAAF1966
  • Academic year2025/2026
  • CourseStatistical Sciences
  • CurriculumDemografico sociale
  • Year1st year
  • Semester1st semester
  • CFU3