Bayes' theorem
1
7.3
- noun
- /ˈbeɪz ðɪərəm/
- Specialized
:
teorema de Bayes
- Most medical decision aids use one of three reasoning methods: Bayes' theorem, decision theory, or symbolic reasoning or "artificial intelligence" techniques.
Examples
-
The posterior distribution of effect size is established using Bayes' theorem.
Academic text (1990) -
A frequentist analysis could certainly include the use of Bayes' theorem.
Blog text (27) -
I use it here as an example of how one can be a user of Bayes' theorem and still have frequentist goals.
Blog text (27) -
Using Bayes' theorem, prior information is combined with new data to obtain updated or posterior information.
Academic text (2010) -
By probabilistic, I mean an approach that will be based on Bayes' theorem and statistical principles that are needed.
Blog text (24) -
In others, he loves to use Bayes' theorem, but his goals are overtly frequentist.
Blog text (27) -
My point is that using Bayes' theorem is neither necessary nor sufficient for defining Bayesian inference.
Blog text (27) -
Decision theory is an extension to Bayes' theorem that lets us add weightings representing the value of each possible outcome to a decision tree to calculate the "expected utility."
Academic text (1991) -
And conversely, it is possible to do Bayesian inference without using Bayes' theorem (as Michael Goldstein, for example, has shown).
Blog text (27) -
Using Bayes' theorem, statisticians can update probabilities as new evidence emerges.
Surface Forms
Morphology
Bayes' + theorem
The phrase is compositionally clear as a possessive named theorem — a B1 learner who knows 'Bayes' (as a name) and 'theorem' will understand it denotes a theorem associated with Bayes. However, the specific probabilistic content (how to compute causes from evidence) is technical and cannot be derived from the constituents alone, so the expression is only partially transparent and requires domain knowledge.
Etymology
Bayes' theorem comes from Thomas Bayes, who showed a simple way to change a guess about what caused something when you get new evidence. So today the theorem is used to 'update' the 'chance' of a cause when we have new information.