Survival Analysis: Statistical Models and Machine Learning algorithms.
Date:
June 29 to July 3. MORNING: 9 to 12h.
Instructor
Ambra Macis
Ambra Macis, PhD in Analytics for Economics and Management (scientific area: Statistics), is Assistant Professor of Statistics at the Department of Economics and Management, University of Brescia.
Her research spans both methodological and applied aspects of Statistical Sciences. She has authored/co-authored several scientific papers in international journals and has presented at national and international conferences.
She teaches undergraduate and graduate courses in the field of Statistics.
She is a member of the BDsports project (bdsports.unibs.it) and the S-training group (https://s-training.eu/), and serves as Associate Editor of the Journal of Sport Analytics.
https://www.unibs.it/it/ugov/person/184884
Language
English
Description
This course offers a solid introduction to survival analysis, covering fundamental concepts such as time-to-event data, censoring mechanisms, and classical statistical models. Building on these foundations, the program explores advanced topics including frailty models and modern machine learning approaches for survival data, such as survival trees and random survival forests. The course combines theoretical concepts with practical applications, giving participants hands-on experience with real-world datasets and analytical tools.
Course goals
By the end of the course, participants will be able to understand and apply key methods of survival analysis, interpret results from both classical and machine learning models, and implement practical solutions using statistical software. The course aims to equip students with the skills needed to analyse time-to-event data in research and applied settings, and to critically evaluate the suitability of different modelling approaches.
Course contents
- Introduction to survival analysis
- Non-parametric methods for survival data
- Parametric survival models
- Semi-parametric models
- Extensions to complex data: frailty models
- Machine learning approaches for survival data: survival trees and random survival forests
All topics will be covered from both theoretical and practical points of view.
Prerequisites
Basic knowledge of statistical modeling is required. Moreover, familiarity of R and RStudio is desirable.
Targeted at
Master of Science and PhD students in STEM subjects. Bachelor’s students are also welcome, provided that they meet the required prerequisites. In addition, post-graduates with an appropriate background who are interested in the topic are welcome to participate.
Evaluation
A final test, covering both theoretical concepts and practical applications, will be administered to assess students’ comprehension of the topics covered throughout the course.
Teaching Methodology and Activities
The course combines theoretical lectures with hands-on sessions in R, allowing students to immediately apply the concepts learned. Practical exercises using real-world datasets will also be provided, encouraging active learning and the development of analytical skills. Students will work on model fitting, performance evaluation, and interpretation of results, ensuring a strong connection between statistical theory and practical application.
Software requirements
R (https://www.r-project.org/) and RStudio (https://posit.co/download/rstudio-desktop/) will be used for practical sessions. Required R packages will be communicated to participants before the course.
Share: