teacup_cerberus: Teacup Cerberus Model

In jsilve24/TeacupCerberus: Bayesian Estimation of Covariation Between and Within Two Multivariate Count Datasets

Description

Estimates Covariance betweeen and withing two datasets accounting for multinomial count uncertainty/error. Returns covariance with respect to CLR coordinates, this can easily be converted to alternative representations. See vignette for details.

Usage

teacup_cerberus(Y1, Y2, alpha1 = NULL, alpha2 = NULL,
  n_samples = 2000)

Arguments

Y1	count data (D1 x N) (e.g., taxa x samples)
Y2	count data (D2 x N) (e.g., food x samples)
alpha1	D1-vector prior for Dirichlet for Y1 (think of it as a "pseudo-count" like thing, must be greater than zero) default: rep(1, D1)
alpha2	D2-vector prior for Dirichlet for Y2 (think of it as a "pseudo-count" like thing, must be greater than zero) default: rep(1, D2)

Details

This fits the following model

Y_1 ~ Multinomial(π_1)

Y_2 ~ Multinomial(π_1)

π_1 ~ Dirichlet(α_1)

π_2 ~ Dirichlet(α_2)

and then transforming posterior samples of that model via

η_1 = CLR^{-1}(η_1)

η_2 = CLR^{-1}(η_2)

and then the s-th sample of Sigma (as a correlation matrix is given by)

Σ^s = cov(cbind(η_1^s, η_2^s))

Value

Array Sigma of dimension (D1+D2) x (D1+D2) x n_samples (Sample of Covariance Matricies)

jsilve24/TeacupCerberus documentation built on Nov. 4, 2019, 3:25 p.m.