Bayesian Statistics with R
The Bayesian framework represents an alternative approach to statistical modelling. Bayesian analysis describes how to update our initially incomplete "prior" knowledge with experimental observations (the "evidence") to obtain a better understanding (the "posterior") of the phenomenon studied. This concept fits very well the way how scientific research works.
In addition, Bayesian Networks represent an important step towards analysing causal relationships which traditional statistical methods cannot model.
This course teaches you the principles of Bayesian statistics with hands-on practical examples in R.
Instructor: András Aszódi.
Topics
- Diagnostic kits, or: Calculating the probability of being sick if you tested positive, using Bayes' Theorem.
- Life and death of a mouse, or: Estimating the probability parameter of the Binomial distribution. Conjugate priors, updating the posterior.
- Blood group phenotypes: estimating several parameters at once.
- Blood group genotypes: modelling complex distributions with Bayesian MCMC techniques.
- Towards modelling causality: Introduction to Bayesian Networks.
Exercises
Online exercises are available when this course is running. Please select the option "Bayes" from the dropdown in the "Request an exercise notebook" form.
Out of scope
We cannot go into the specific data analysis problems of your particular project.
Note that "Bayesian statistics" is a huge subject and this introductory course only teaches you the basics.
Prerequisites
Good knowledge of how to calculate probabilities and familiarity with discrete and continuous distributions are required. If you attended our Think Statistics! with R training then you are well prepared.
In addition, some basic familiarity with R is required. Our R as a programming language training provides a good foundation.
Practical information
Number of participants: minimum 5, maximum 10.
Length: The course takes one half-day, from 09:00 to 13:00 with 2 breaks.