transform_stm: Refits a STM with a content-based covariate so that all...

In reaganmozer/textmatch: Toolkit for Matching Textual Data and Evaluating Textual Similarity

Refits a STM with a content-based covariate so that all document-level topic-proportions are estimated "as-treated". Also allows for calculation of the SR sufficient reduction and optional coarsening to reduce the dimension of the data, reduce noise, or apply other design rules (e.g. - to exclude words that occur in too few or too many documents).

Description

Refits a STM with a content-based covariate so that all document-level topic-proportions are estimated "as-treated". Also allows for calculation of the SR sufficient reduction and optional coarsening to reduce the dimension of the data, reduce noise, or apply other design rules (e.g. - to exclude words that occur in too few or too many documents).

Usage

transform_stm(mod, out, Z, calc.SR = FALSE, coarsen = FALSE, simplex = FALSE)

Arguments

mod	a fitted stm object
out	the original call to the STM
Z	an indicator for treatment assignment
calc.SR	an indicator for returning the sufficient reduction. Default is TRUE.
coarsen	an indicator for returning the coarsened STM

Value

A bounded DFM

reaganmozer/textmatch documentation built on March 7, 2024, 2:41 p.m.