Παρουσίαση/Προβολή

(STAT354) -  ΝΙΚΟΛΑΟΣ ΔΕΜΙΡΗΣ

Περιγραφή Μαθήματος

This course is offered in the MSc in Statistics.

 

For the spring semester of the academic year 2024-25 the lectures take place at room 609 every Friday 12-3.

 

At the end of the course students will have knowledge of the basic principles, methods and implementation tools of the main data science techniques that are being used in the analysis of health data.

 

The course has the following content

Basic concepts in survival analysis, definitions, hazard and survival functions, relationships, parametric methods, likelihood function, Exponential and Weibull Models, applications in R

Non-parametric methods: Kaplan-Meier estimator, Greenwood and Nelson-Aalen estimator, graphical goodness of fit, log rank test.

Regression models, Cox proportional hazards, Survival Analysis theory, counting processes, applications in R

Martingale/Deviance/Schoenfeld residuals. Heterogeneity and frailty models, LASSO and elastic net, hyperparameter selection via cross-validation, applications in glmnet

Non-proportional hazards models, additive hazards, accelerated failure time, proportional odds, competing risks and (non-)identifiability, multi-state models

Prospective and retrospective studies, (non)interventional, AR, RR and OR, equivalence of OR. Screening tests, PPV/NPV and sensitivity/specificity

Clinical trial design and analysis, protocol, sample size calculations, phase I, MTD, 3+3 design, Phase II, safety and efficacy, phase III. Real world vs (and/or) randomised data.

CRM+adaptive designs, Simon 2-stage design, Bayesian and historical/synthetic controls

Meta analysis, systematic reviews, fixed effects, heterogeneity, random effects, publication bias, funnel plots, indirect treatment comparisons and network meta analysis, example applications in health economics using ICER, INB and CEAC.

Evidence synthesis and conflict diagnostics

Introduction to Epidemic models, main results, vaccination and control.

Basic stochastic models, branching processes and coupling, functional LLN and CLT, connections between the different types of model.

Inference for chain binomial models using MCMC. Inference for deterministic models using HMC.

Heterogeneity, multiple age-groups, contact matrices, epidemics among households.

Epidemics on networks

 

No single book covers the necessary material, some indicative books are the following

Cox and Oakes (1984) Analysis of Survival Data, Chapman and Hall

Klein and Moeschberger (1997) Survival Analysis: Techniques for Censored and Truncated Data. Springer Series in Statistics for Biology and Health

Hosmer, Lemeshow and May (2008) Applied Survival Analysis: Regression Modeling of Time-to-Event Data, Wiley

Andersen, Borgan, Gill and Keiding (1993) Statistical Models Based on Counting Processes, Springer

Rosner (2015) Fundamentals of Biostatistics, Cengage

Andersson and Britton (2000) Stochastic epidemic models and their statistical analysis. Springer Lecture Notes in Statistics.

Diekmann O., Heesterbeek, J.A.P. and Britton, T. (2013). Mathematical tools for understanding infectious disease dynamics. Princeton University Press.

 

Ημερομηνία δημιουργίας

Τρίτη, 13 Φεβρουαρίου 2024