The prior information of the unknown parameter x is represented by a distribution form , which is called the prior distribution of the unknown parameter .
The probability that the result is a cause is the posterior probability.
Likelihood estimation is based on the reason to speculate the probability that the cause will lead to the result.
: Indicates the observed data (result)
: The parameters that determine the data distribution (reasons)
: Prior : posterior posterior probability
: Likelihood estimation
: evidence Probability statistical information about x.
MLE maximum likelihood estimation:
Maximum posterior estimation:
When the maximum posterior and maximum likelihood are optimized, there is a prior term at the time of the maximum posterior .