Description: There is a long history of the application of statistical methods on collected data (for example data from experiments). The current replication crisis in science is evidence that sometimes (maybe even often) statistical methods are used wrongly or their results misinterpreted. In this course we give a coherent introduction to statistical analyses and their interpretation that can help avoid these problems.

In this course we introduce a statistical mindset ? a way of thinking: a statistical paradigm of evidence interpretation. We will focus on conditional probabilities: is the question the patient is asking ?what is the probability of having a positive test result given I am sick?? or is it ?what is the probability of being sick given a positive test result??. We answer this question ? and discuss why the two questions are indeed very different questions.    We proceed to analysing data with probabilistic modelling. We do that by first discussing data-generating processes and causal models, including using these to identify the correct variables to control for in the statistical analysis. Then we proceed to linear regression (a simple linear regression and a linear regression with both continuous and qualitative explanatory variables including potentially interaction effects). We potentially also consider other types of models (for example logistic regression and/or correlated measurements).

We will use R (https://www.r-project.org) and the Stan software for Bayesian inference (https://mc-stan.org).

Prerequisites: Working knowledge of R. (For example as obtained through the PhD course ?Data Science Using R?.)

Learning objectives: Analysing data using statistical methods ensuring reproducible research findings, including  probabilistic modelling and using software for Bayesian inference.

Teaching methods: Oral presentations, exercises, hand-in.

Criteria for assessment: The students need to participate actively and they will get a written hand-in assignment that must be approved to pass the course.

Key literature: The course does not follow any particular text book, but rather the course slides and supplied notes will be the main course material.

Organizer: Mikkel Meyer Andersen

Lecturers: Mikkel Meyer Andersen

ECTS: 2.0

Time: 17 April and 1 May 2025

Place: Aalborg University

Zip code: 9220

City: Aalborg

Maximal number of participants: 40

Deadline: 27 March 2025

Disclaimer:
DDSA has explicit permission from Arcanic and the owners of the https://phdcourses.dk/ website to display the courses on ddsa.dk.