You are here:
Start
Department of Statistics
Research
Bayesian Inference
Bayesian Inference

Bayesian Inference

Bayesian inference grew up and was the dominant form of statistical inference in the 19th century and the beginning of the 20th century, with known representatives such as Laplace, Borel and Keynes.

In the 1920s began a period of so-called Neyman-Pearson inference. Bayesian inference re-emerged over the last 20-25 years and is now widely accepted.

Bayesian inference is based on simple principles - that conclusions must be logically consistent; that conclusions shall be based on what has been observed or what is already known; and that the results can be used as a basis for decision.

Since conclusions are to be based on everything that is known, two individuals with different background knowledge may draw different conclusions from the same experiment.

If an investigation or experiment is to be scientifically proven the data must be so extensive as to convince even individuals who, prior to the experiment, harbour reasonable doubt, given such individuals are open to rational debate.

Staff

Mattias Villani, professor

Researchers

Andriy Andreev, senior lecturer
Raul Cano, senior lecturer
Jessica Franzén, Senior lecturer
Gebrenegus Ghilagaber, professor
Oskar Gustafsson, PhD student
Parfait Munezero, PhD student
Oscar Oelrich, PhD student

Last updated: October 4, 2018
Page editor: Håkan Slättman
Source: Department of Statistics

Tell a friend

Contacts

Deputy Head of Department
Mattias Villani

Contacts