This issue aims to track the extention of the PointEstimateModel to multivatiate distributions.
Let X_1, X_2, ..., X_K distributions, they are represented by x_{1,1}, ..., x_{k, n}, ..., x_{K, N}, where k \in 1,...,K and n \in 1,...,N using gaussian quadrature.
Each x_{k,n}= \mu_k + \epsilon_{k,i} \sigma_k
The model implemented in this repo is this one and applies for a single variable.
The extension to multivariate is also discussed in this paper
This issue aims to track the extention of the PointEstimateModel to multivatiate distributions.
Let X_1, X_2, ..., X_K distributions, they are represented by x_{1,1}, ..., x_{k, n}, ..., x_{K, N}, where k \in 1,...,K and n \in 1,...,N using gaussian quadrature.
Each x_{k,n}= \mu_k + \epsilon_{k,i} \sigma_k
The model implemented in this repo is this one and applies for a single variable.
The extension to multivariate is also discussed in this paper