Multi Model Probability Distribution is a model comprised of an unspecified combination of multiple probability distribution functions (univariate and/or multivariate)
Definition
on each trial, the choice of which probability distribution generates the sample is determined by sampling a distribution identity from a multinoulli distribution
- 𝐏(𝑥) = 𝛴𝑖𝐏(𝑐=𝑖)𝐏(𝑥|𝑐=𝑖)
where:
- 𝐏(𝑐) - is the multinoulli distribution over distribution identities
Example
a mixed distribution modeled with 3 normal distributions
