Matteo Tanadini: Catalogue data in Autumn Semester 2019

Name Dr. Matteo Tanadini
Seminar für Statistik (SfS)
ETH Zürich, HG G 14.1
Rämistrasse 101
8092 Zürich

401-0629-00LApplied Biostatistics4 credits3GM. Tanadini
AbstractThis course covers the main methods used in Biostatistics. It starts by revising Linear Models (Regression, Anova), then moves to Generalised Linear Models (logistic regression and methods for count data) and finally introduces more advanced topics (Linear Mixed-Effects Models and Survival Analysis). The course strongly focuses on applied aspects of data analysis.
ObjectiveAfter this course students:
- revised Linear Models
- revised or got introduced to Generalised Linear Models
- got introduced to Linear Mixed-Effects Models and Survival Analysis
- are able to select among these methods to solve an applied problem in Biostatistics
- can perform the analysis using the statistical software R
- can interpret the results of such an analysis and draw valid "biological" conclusions
ContentThis course is structured into three parts. The first part focuses on Linear and Generalised Linear Models. The second part introduces more advanced methodologies such as Linear Mixed-Effects Models and Survival Analysis. Both, part one and two will included the following topics: exploratory data analysis, model fitting, model "selection", residual diagnostics, model validation and results interpretation. Analyses will be carried out by using the statistical software R. Finally, in the third part of the course students will be analysing real-world data sets to put into practice the knowledge and skills acquired during the first two parts.
Prerequisites / NoticeThe statistical software R will be used in the exercises. If you are unfamiliar with R, it is highly recommend to view the online R course etutoR.
447-0649-01LApplied Statistical Regression I Restricted registration - show details
Only for DAS and CAS in Applied Statistics.
4 credits1V + 1UM. Tanadini
AbstractSimple and multiple regression models, with emphasis on practical aspects and interpretation of results, analysis of residuals and model selection.
ObjectiveUnderstanding the multiple linear regression model and its importance for modelling and prediction. Practice of regression analyses using the statistical software R and correct interpretation of results. Model critique by analysis of residuals. Strategies for model selection.