Estimating text regressions using txtreg_train

• Autores: Carlo Schwarz
• Localización: The Stata journal, ISSN 1536-867X, Vol. 23, Nº. 3, 2023, págs. 799-812
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • In this article, I introduce new commands to estimate text regres- sions for continuous, binary, and categorical variables based on text strings. The command txtreg_train automatically handles text cleaning, tokenization, model training, and cross-validation for lasso, ridge, elastic-net, and regularized logis- tic regressions. The txtreg_predict command obtains the predictions from the trained text regression model. Furthermore, the txtreg_analyze command facil- itates the analysis of the coefficients of the text regression model. Together, these commands provide a convenient toolbox for researchers to train text regressions.

    They also allow sharing of pretrained text regression models with other researchers.