Regression Analysis and Forecasting Using Textual Data from a Time-Varying Dictionary

Provides functionalities based on the paper "Time Varying Dictionary and the Predictive Power of FED Minutes" (Lima, 2018) . It selects the most predictive terms, that we call time-varying dictionary using unsupervised machine learning techniques as lasso and elastic net.

Travis build status AppVeyor build status Coverage Status

The goal of TextForecast is to carry out forecasting and regression analysis using textual analysis and supervised machine learning techniques as LASSO, Elastic Net and Ridge Regression to select the most predictive words/terms.


You can install the released version of TextForecast from github with:



Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.1.0 by Lucas Godeiro, 14 days ago

Report a bug at

Browse source code at

Authors: Luiz Renato Lima [aut] , Lucas Godeiro [aut, cre]

Documentation:   PDF Manual  

GPL-3 license

Imports SnowballC, forecast, rpart, stats, text2vec, tidyr, tidytext, tm, tsDyn, wordcloud, dplyr, plyr, udpipe, class, lars, tau, RColorBrewer, forcats, ggplot2, glmnet, pdftools, parallel, doParallel

Suggests knitr, rmarkdown, covr

See at CRAN