Package: FactorHet 1.0.0

FactorHet: Estimate Heterogeneous Effects in Factorial Experiments Using Grouping and Sparsity

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 (2024) <doi:10.48550/ARXIV.2201.01357> provide further details.

Authors:Max Goplerud [aut, cre], Nicole E. Pashley [aut], Kosuke Imai [aut]

FactorHet_1.0.0.tar.gz
FactorHet_1.0.0.zip(r-4.5)FactorHet_1.0.0.zip(r-4.4)FactorHet_1.0.0.zip(r-4.3)
FactorHet_1.0.0.tgz(r-4.4-x86_64)FactorHet_1.0.0.tgz(r-4.4-arm64)FactorHet_1.0.0.tgz(r-4.3-x86_64)FactorHet_1.0.0.tgz(r-4.3-arm64)
FactorHet_1.0.0.tar.gz(r-4.5-noble)FactorHet_1.0.0.tar.gz(r-4.4-noble)
FactorHet_1.0.0.tgz(r-4.4-emscripten)FactorHet_1.0.0.tgz(r-4.3-emscripten)
FactorHet.pdf |FactorHet.html✨
FactorHet/json (API)

# Install 'FactorHet' in R:

install.packages('FactorHet', repos = c('https://mgoplerud.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/mgoplerud/factorhet/issues

Uses libs:

c++– GNU Standard C++ Library v3

Datasets:

immigration - Small dataset on immigration preferences

On CRAN:

cpp

3.18 score 3 stars 1 scripts 22 exports 46 dependencies

Last updated 4 days agofrom:eaa6e84cb0. Checks:9 OK. Indexed: yes.

Target	Result	Latest binary
Doc / Vignettes	OK	Jan 14 2025
R-4.5-win-x86_64	OK	Jan 14 2025
R-4.5-linux-x86_64	OK	Jan 14 2025
R-4.4-win-x86_64	OK	Jan 14 2025
R-4.4-mac-x86_64	OK	Jan 14 2025
R-4.4-mac-aarch64	OK	Jan 14 2025
R-4.3-win-x86_64	OK	Jan 14 2025
R-4.3-mac-x86_64	OK	Jan 14 2025
R-4.3-mac-aarch64	OK	Jan 14 2025

Exports:ACE AME AMIE cjoint_plot diff_AME FactorHet FactorHet_control FactorHet_init FactorHet_mbo FactorHet_mbo_control FactorHet_refit HTE_by_individual HTE_by_moderator manual_AME margeff_moderators marginal_ACE marginal_AME marginal_AMIE moderator_AME posterior_by_moderators posterior_FactorHet visualize_MBO

Dependencies:backports BBmisc checkmate cli colorspace data.table fansi farver fastmatch ggplot2 glue gtable isoband labeling lattice lbfgs lhs lifecycle magrittr MASS Matrix mgcv mlr mlrMBO munsell nlme parallelMap ParamHelpers pillar pkgconfig R6 RColorBrewer Rcpp RcppArmadillo RcppEigen rlang scales smoof stringi survival tibble utf8 vctrs viridisLite withr XML

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Calculate marginal effects	ACE AME AMIE manual_AME
Plot a FactorHet object	cjoint_plot
Difference between AMEs in each group	diff_AME
Estimate heterogeneous effects in factorial and conjoint experiments	FactorHet FactorHet_mbo
Control for FactorHet estimation	FactorHet_control
Arguments for initializing FactorHet	FactorHet_init
Control for model-based optimization	FactorHet_mbo_control
Refit model using estimated sparsity patterns	FactorHet_refit
Generic methods for FactorHet models	AIC.FactorHet BIC.FactorHet coef.FactorHet FactorHet-class formula.FactorHet logLik.FactorHet plot.FactorHet posterior_FactorHet print.FactorHet print.FactorHet_vis summary.FactorHet vcov.FactorHet visualize_MBO
Estimate heterogeneous treatment effects by individual or moderator	HTE HTE_by_individual HTE_by_moderator
Small dataset on immigration preferences	immigration
Compute association between moderators and group membership	margeff_moderators
Visualize the posterior by observed moderators	posterior_by_moderators
Predict after using FactorHet	predict.FactorHet