Causal inference for proportional outcomes with a point mass at a boundary (zero, one, or both), via the zero-or-one inflated beta (BIc) regression of Ospina and Ferrari (2012). causalbic estimates the average and conditional treatment effect by g-computation and decomposes it exactly into a discrete-component contribution (through the boundary probabilities) and a continuous-component contribution (through the conditional mean), a consequence of the block-diagonal Fisher information of the BIc likelihood. A diagnostic layer built on the same separability adds a causal conformal normal curvature with the same additive decomposition, a per-unit causal influence screen, and a Hölder-duality sensitivity bound for unmeasured confounding.
Companion package to Ospina (2026), Causal inference for proportional outcomes via likelihood displacement in inflated beta regression, submitted to The Annals of Applied Statistics. This repository is the package only; it does not include the manuscript or the scripts that reproduce its simulation and application.
Installation
# install.packages("pak")
pak::pak("Raydonal/causalbic")Example
library(causalbic)
set.seed(1)
dat <- simulate_bic(600)
fit <- bic_fit(y ~ T + W2, nu.formula = ~ T + W1, data = dat)
#> GAMLSS-RS iteration 1: Global Deviance = 448.3807
#> GAMLSS-RS iteration 2: Global Deviance = 371.5585
#> GAMLSS-RS iteration 3: Global Deviance = 367.9107
#> GAMLSS-RS iteration 4: Global Deviance = 367.8287
#> GAMLSS-RS iteration 5: Global Deviance = 367.8272
#> GAMLSS-RS iteration 6: Global Deviance = 367.8272
bic_gcomp(fit, treat = "T", data = dat)
#> tau tau_alpha tau_mu
#> 0.06570652 -0.07816620 0.14387272The decomposition is exact, not approximate:
g <- bic_gcomp(fit, "T", dat)
g["tau_alpha"] + g["tau_mu"] - g["tau"]
#> tau_alpha
#> 0See vignette("causalbic") for the diagnostic layer (curvature, per-unit influence, sensitivity to unmeasured confounding).