ABSTRACT
In this chapter, the authors introduce tidymodels, a collection of packages for modeling and Machine Learning (ML) using tidyverse principles. Tidymodels comes with a handy workflow for all sorts of typical prediction tasks. They implement and showcase the entire cycle from model specification, training, and forecast evaluation within the tidymodels universe. The authors illustrate that stock characteristics such as size provide valuable pricing information in addition to the market beta. Kuhn and Silge provide a thorough introduction into all tidymodels components. Glmnet as developed and released in sync with Tibshirani and provides an R implementation of Elastic Net estimation. The tidymodels workflow encompasses the main stages of the modeling process: pre-processing of data, model fitting, and post-processing of results.
