Sometimes, we want a model that generates a range of possible outcomes around each prediction. Other times, we just care about point predictions and may opt to use a fancy model like XGBoost. But what if we want the best of both worlds: getting a range of predictions while still using a fancy model? That’s where bootstrapping comes to the rescue! By using bootstrap resampling, we can create many models that produce a prediction distribution – regardless of the model type! In this talk, I’ll give an overview of bootstrap resampling for prediction, the pros/cons of this method, and how to implement it as a part of a tidymodel workflow with the workboots package.
Talk materials are available at https://github.com/markjrieke/rstudio-conf-2022.
I am a senior consumer experience (CX) analyst at Memorial Hermann Health System where I use R and tidymodels to provide actionable insights from patient satisfaction survey data. I love making beautiful charts, working on home improvement projects, and playing jazzy piano. I live in Houston, TX, with my fiancé and two obnoxious yet lovable pets.
