A dataset used for modeling support for gay marriage in the United States, 
combining individual- and state-level covariates from a 2006 survey.
<ul>
<li>A data frame with 5000 rows and 11 variables:<dl>
 <dt>id</dt>
<dd>Unique observation identifier</dd> <dt>state</dt>
<dd>Two-letter abbreviation for U.S. state</dd> <dt>stateid</dt>
<dd>Numeric identifier for the state</dd> <dt>region</dt>
<dd>Region code</dd> <dt>age</dt>
<dd>Age group (1 = 18-30, 2 = 31-50, 3 = 51-65, 4 = 65+)</dd> <dt>gXr</dt>
<dd>Gender and race interaction</dd> <dt>educ</dt>
<dd>Education level (1 = LTHS,2 = HS,3 = Some Coll,4 = Coll+)</dd> <dt>supp_gaymar</dt>
<dd>Support for gay marriage (0 = oppose, 1 = support)</dd> <dt>pvote</dt>
<dd>Republican presidential vote share in the previous election</dd> <dt>religcon</dt>
<dd>Proportion of population identifying as religious conservatives</dd> <dt>libcon</dt>
<dd>State-level ideology score (liberal to conservative)</dd>
</dl></li>
</ul>

datasets

Complex machine learning models are often difficult to interpret. Shapley values serve as a powerful tool to understand and explain why a model makes a particular prediction. This package computes variable contributions using permutation-based Shapley values for Bayesian Additive Regression Trees (BART) and its extension with Post-Stratification (BARP). The permutation-based SHAP method proposed by Strumbel and Kononenko (2014) <doi:10.1007/s10115-013-0679-x> is grounded in data obtained via MCMC sampling. Similar to the BART model introduced by Chipman, George, and McCulloch (2010) <doi:10.1214/09-AOAS285>, this package leverages Bayesian posterior samples generated during model estimation, allowing variable contributions to be computed without requiring additional sampling. The BART model is designed to work with the following R packages: 'BART' <doi:10.18637/jss.v097.i01>, 'bartMachine' <doi:10.18637/jss.v070.i04>, and 'dbarts' <https://CRAN.R-project.org/package=dbarts>. For XGBoost and baseline adjustments, the approach by Lundberg et al. (2020) <doi:10.1038/s42256-019-0138-9> is also considered. The BARP model proposed by Bisbee (2019) <doi:10.1017/S0003055419000480> was implemented with reference to <https://github.com/jbisbee1/BARP> and is designed to work with modified functions based on that implementation. BARP extends post-stratification by computing variable contributions within each stratum defined by stratifying variables. The resulting Shapley values are visualized through both global and local explanation methods.

Dong-eun Lee

bartXViz

Visualization of BART and BARP using SHAP

Eun-Kyung Lee

svy function

A dataset used for modeling support for gay marriage in the United States, 
combining individual- and state-level covariates from a 2006 survey.
<ul>
<li>A data frame with 5000 rows and 11 variables:<dl>
 <dt>id</dt>
<dd>Unique observation identifier</dd>

 <dt>state</dt>
<dd>Two-letter abbreviation for U.S. state</dd>

 <dt>stateid</dt>
<dd>Numeric identifier for the state</dd>

 <dt>region</dt>
<dd>Region code</dd>

 <dt>age</dt>
<dd>Age group (1 = 18-30, 2 = 31-50, 3 = 51-65, 4 = 65+)</dd>

 <dt>gXr</dt>
<dd>Gender and race interaction</dd>

 <dt>educ</dt>
<dd>Education level (1 = LTHS,2 = HS,3 = Some Coll,4 = Coll+)</dd>

 <dt>supp_gaymar</dt>
<dd>Support for gay marriage (0 = oppose, 1 = support)</dd>

 <dt>pvote</dt>
<dd>Republican presidential vote share in the previous election</dd>

 <dt>religcon</dt>
<dd>Proportion of population identifying as religious conservatives</dd>

 <dt>libcon</dt>
<dd>State-level ideology score (liberal to conservative)</dd>


</dl></li>
</ul>

svy: Survey Data on Support for Gay Marriage (2006)

Description

Arguments

References