Prediction, classification and causal inference modelling in health data
Mon, May 15
|Online (Zoom) or in Presence
Time & Location
May 15, 2023, 10:00 AM – May 25, 2023, 5:00 PM
Online (Zoom) or in Presence
Guests
Info & Program Details
When
15, 16, 18, 22, 23, 25 May 2023 (three hours per day)
Price
150 euros students (including PhD, and post-graduate students)
250 euros for academic staff
350 euros for non-academic staff
(IVA esente ai sensi dell’art. 10 DPR 633/72).
Deadline for registration: 10th May 2023
Schedule
- Monday (15 May) h 10-13: Probability and conditional independence
Lecture (2 hours)
Probability, random variables, binomial distribution, normal distribution, hypothesis testing, conditional independence and direct acyclic graphs
Practical session (1 hour)
Introduction to R, data simulation and visualization
- Tuesday (16 May) h 10-13: Regression modelling and dimension reduction for prediction and classification
Lecture (2 hours)
Linear regression, logistic regression, model selection, principal component analysis
Practical session (1 hour)
Introduction to R, data simulation and visualization
- Thursday (18 May) h 10-13: Practical session and Tutorial
Practical session (2 hours)
Data analysis in R on regression modelling and dimension reduction
Tutorial (1 hour)
Recap of course materials in Week 1, Q&A session
- Monday (22 May) h 14-17: Machine learning methods for prediction, classification and model evaluation
Lecture (2 hours)
Random forest, support vector machine, XGBoost, confusion matrix
Practical session (1 hour)
Data analysis in R on prediction, classification and model evaluation using machine learning methods
- Tuesday (23 May) h 14-17: Causal inference
Lecture (2 hours)
Confounding, propensity score, Mendelian randomization
Practical session (1 hour)
Data analysis in R on prediction, classification and model evaluation using machine learning methods
- Thursday (25 May) h 14-17: Practical session and Seminar talk
Practical session (2 hours)
Data analysis in R on propensity score and Mendelian randomization analysis
Seminar talk (1 hour)
Bayesian Mendelian randomization and causal networks
Where?
ONLINE, via Zoom platform. Students can also attend the course in presence at University of Pavia, Cascina Cravino, Via Bassi 21, 27100 Pavia Italy.
Teacher
Dr. Hui Guo. University of Manchester
Hui Guo leads the Centre for Biostatistics at the University of Manchester where she has worked as a Lecturer and then a Senior Lecturer in Biostatistics since 2014. Hui serves on the University’s management board of the Institute for Data Science and Artificial Intelligence.
She has been a Fellow at The Alan Turing Institute (UK) since 2018. She completed a PhD in Statistics and worked as a Research Associate at the University of Cambridge. Prior to this, she obtained her MSc and PG Diploma in Statistics at UCL. Hui’s research focuses on causal
inference in health-related observational studies. Her current work is on exploring causal mechanisms underlying clinical outcomes from large-scale genetic association data. She is also interested in genetic and/or biological causal pathways and networks of certain diseases.
Billing
After registering for the course, interested parties will be contacted by the administrative secretariat who will explain the payment methods. The actual registration will be finalized with the payment of the registration fee.
Organizational secretariat:
Dott.ssa Gianfranca Corbellini
Phone: 0382 987526
E-mail: dbbs.master@unipv.it
Per info contattare anche:
Dott.ssa Teresa Fazia
teresa.fazia@unipv.it