The Structural Topic Model (STM) allows researchers to estimate topic models with document-level covariates. The package also includes tools for model selection, visualization, and estimation of topic-covariate regressions. Methods developed in Roberts et al (2014) <doi:10.1111/ajps.12103> and Roberts et al (2016) <doi:10.1080/01621459.2016.1141684>. Vignette is Roberts et al (2019) <doi:10.18637/jss.v091.i02>.
copied from cf-post-staging / r-stm