Estimates heterogeneous effects in factorial (and conjoint)
models. The methodology employs a Bayesian finite mixture of
regularized logistic regressions, where moderators can affect each
observation's probability of group membership and a sparsity-inducing
prior fuses together levels of each factor while respecting
ANOVA-style sum-to-zero constraints. Goplerud, Imai, and Pashley
(2025) <doi:10.1214/24-AOAS1994> provide further details.
| Version: |
1.0.1 |
| Depends: |
R (≥ 3.4.0) |
| Imports: |
Rcpp (≥ 1.0.1), Matrix, ggplot2, ParamHelpers, mlr, mlrMBO, smoof, lbfgs, methods, utils, stats |
| LinkingTo: |
Rcpp, RcppEigen (≥ 0.3.3.4.0) |
| Suggests: |
FNN, RSpectra, mclust, ranger, tgp, testthat, covr, tictoc |
| Published: |
2026-11-04 |
| DOI: |
10.32614/CRAN.package.FactorHet |
| Author: |
Max Goplerud [aut, cre],
Nicole E. Pashley [aut],
Kosuke Imai [aut] |
| Maintainer: |
Max Goplerud <mgoplerud at austin.utexas.edu> |
| BugReports: |
https://github.com/mgoplerud/FactorHet/issues |
| License: |
GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
| URL: |
https://github.com/mgoplerud/FactorHet |
| NeedsCompilation: |
yes |
| Materials: |
README, NEWS |
| CRAN checks: |
FactorHet results |