We recommend using dirichlet_multinomial_glm_anova instead, but this is here for comparison.

betaBinomialGLM(
  ys,
  ns,
  xFull,
  xNull,
  concShape = 1.0001,
  concRate = 1e-04,
  ...
)

Arguments

ys

numerator counts

ns

denominator counts

xFull

matrix of covariates for full model. First column must be ones.

xNull

matrix of covariates for null model. First column must be ones.

concShape

Gamma shape parameter for concentration parameter

concRate

Gamma rate parameter for concentration parameter

...

will be passed on the rstan::optimizing, so can be used for example to set the algorithm used (default is LBFGS).