Skip to contents

Families for paired counts

Constructors for the bipois and binegbin families, which model a pair of counts by trivariate reduction with Poisson or negative-binomial components. Each family has a second constructor for partially paired designs, listed below, and the two constructors of a family share a likelihood and the post-processing methods.

bipois() bipois_stanvars() log_lik_bipois() posterior_predict_bipois() posterior_epred_bipois()
Joint bivariate-Poisson custom family for brms
binegbin() binegbin_stanvars() log_lik_binegbin() posterior_predict_binegbin() posterior_epred_binegbin()
Joint bivariate custom family with negative-binomial components for brms

Partially paired designs

Constructors for partially paired designs, in which source 1 has missing values. On an unpaired row, the likelihood term is the marginal probability of y2, so the row informs the shared component and the exclusive component of source 2.

bipois_partialobs() bipois_partialobs_stanvars()
Joint bivariate-Poisson family for partially observed pairs
binegbin_partialobs() binegbin_partialobs_stanvars()
Joint bivariate family with negative-binomial components for partially observed pairs

Prediction and imputation

Draws and expectations of both counts together, and of y1 given the recorded y2. The functions conditional on y2 impute y1 on the unpaired rows of a fit from bipois_partialobs() or binegbin_partialobs().

posterior_predict_y1_and_y2() posterior_epred_y1_and_y2()
Draws and expectations of both counts from the joint model
posterior_predict_y1_given_y2() posterior_epred_y1_given_y2()
Draws and expectations of y1 given y2

Congruence–bias coordinates

Conversion between the native parameters and the congruence–bias coordinates (the midpoint of the two expected counts, the congruence, and the source bias), and the brms formula for a model in the coordinates.

mfd_to_dpars()
Convert from the congruence–bias parameterisation to native parameters
dpars_to_mfd()
Convert from native parameters to the congruence–bias parameterisation
bf_mfd()
Build a brms formula in the congruence–bias parameterisation

Simulation

Random generation of paired counts from the generating model of the bipois and binegbin families, with the shared and source-exclusive components of each pair.

rbicount()
Random generation of paired counts