Teaching

Courses

I am currently teaching three courses at the University of Copenhagen (KU):

• Epidemiological methods in medical research. This is a 10 day Ph.D. course for medical researchers, starting in January and ending in March . I am the course director in collaboration with the epidemiology section of KU.
  • general information: planning, Project description, R refresher, stat refresher.
  • material for day 1: Introduction, lecture 1, lecture 2, R demo.
  • material for day 2: practical 1.

• Statistical analysis of repeated measurements and clustered data. This is 6 day Ph.D. course for medical researchers. It is led by Julie Forman and takes place in May.

• Survival Analysis. This is a 7 week course for Master students in statistics. It is led by Frank Eriksson and takes place in December-January.

Workshops

With Julie Forman, we have made a workshop on linear mixed models (LMMs) for the method week at Karolinska Institutet:

• theoretical part about LMMs and their implementation in LMMstar
• practical part: single group, two groups observational, two groups randomized, optimizer, statistical inference, predictions

For the Brain drug project, I am also creating a workshop on Time-to-event analysis for registry data:

• theoretical part.
• practical part.

Useful and “pedagogical” references

• Adjustment for multiple comparisons: 3 groups Goeman, 2022, general case Dmitrienko, 2013,
• Causality: Hernan, 2004, Pearce 2020
• Competing risks: Andersen, 2012
• DAGs: summary of the DAGs from Hernan and Robin book. Includes DAGs related to measurement error.
• Efron’s paradox dice: Thangavelu 2007
• Groups sequential design: Todd, 2001
• Interaction vs. effect modification: VanderWeele, 2009
• Mann-Whitney parameter: Fay, 2018
• Mediation: continuous outcome & hypotheses Vanderweele, 2009, binary outcome Vanderweele, 2010
• Non-collapsability of odds ratio: Greenland, 2021
• Observed power: Hoenig, 2001
• Per protocol analysis: DAGs showing the bias of naive methods Hernán, 2012 and recommandations Hernán, 2017
• Recurrent events: Furberg, 2021
• Risk, rate, and competing risks: Beyersmann, 2014
• Sample size calculation for existing databases: Hernan, 2022
• Selection bias: Hernandez-Diaz, 2006)
• t-test vs. Mann-Whitney: Skovlund, 2001
• Table 1 (no p-value): STROBE, 2007: “Inferential measures such as standard errors and confidence intervals should not be used to describe the variability of characteristics, and significance tests should be avoided in descriptive tables.”
• Table 2 Fallacy: Westreich, 2013

Reporting guidelines (https://www.equator-network.org/)

• Randomised trials: CONSORT
• Observational studies: STROBE
• Diagnostic and prognostic studies: TRIPOD

Learning R

Basics:

• Introduction tutorial made by colleague from KU and covering installation, data management, and basic data visualisation
• Basic statistic course made by a colleague from KU covering basic notions in statistics and corresponding R code
• basic R cheat sheed or long
• R studio cheat sheed

Efficient data management using data.table:

• Introduction
• Update data table objects
• Wide and long format
• Slide show
• cheat sheet

Efficient generation of graphical displays using ggplot2:

• tutorial
• cheat sheet

Regular expressions:

• help page
• cheat sheet (page 2)

Specialized or advanced topics:

• R markdown tutorial and R markdown cheat sheet for generating documents mixing text, R code, R outputs, and graphical display generated in R.
• Linear models, diagnostics, and remedies
• Multiple imputation workshop and “homemade” summary for using the mice package.
• Analyzing repeated measurements
• formula in R (section 11.1)
• If and do
• Functions
• Simulating data

If you are using R and you think you’re in hell, this is for you.