Mixture model to handle outliers within model

[kaitejohnson]

Problem

We currently have a separate outlier handling step. We probably want to instead incorporate outlier detection directly into the generative model. One option for doing this is using mixture model, as is described here: https://betanalpha.github.io/assets/case_studies/identifying_mixture_models.html https://betanalpha.github.io/assets/case_studies/mixture_models.html

Requirements

• [ ] Implement mixture model for handling outliers from two different generative processes