Bayes Rule
It is a powerful law of probability that brings in the concept of ‘subjectivity’ or ‘the degree of belief’ into the cold, hard statistical modeling.
This is most practical model for decision making. It makes use of priors(knowledge or assumptions or beliefs), confidence levels and when new data is available it helps us update our initial beliefs(priors). As and when we gather more data the confidence levels keeps going up. The probability goes up or down based on whether the newly available data support our priors or counters them.
Ability to recursively apply Bayes rule makes it more interesting. That is a like a meta model that we can use for increasing our confidence levels in other models. meta rules
Further reading
- https://towardsdatascience.com/bayes-rule-with-a-simple-and-practical-example-2bce3d0f4ad0
- https://www.mathsisfun.com/data/bayes-theorem.html
- https://betterexplained.com/articles/an-intuitive-and-short-explanation-of-bayes-theorem/