NoteReviewers Ryan Peterson (2025-11-24). Patrick Breheny (2025-12-12). What connotation do you attach to the word “bias”? A negative one? In this post we will see why not all bias is bad… at least when it comes to building predictive models. In fact, for many years, statisticians have recognized the benefits of biased estimators in reducing prediction error. Perhaps you knew this, but if not, don’t worry.

NoteReviewed by Logan Harris on November 21, 2025 Introduction What are glass-box models? Glass-box models are transparent, intrinsically interpretable alternatives to their opaque counterparts, black box models.

A data-dictated look at whether Denver’s free August public transit policy had its intended effect on air quality. Image credit: National Renewable Energy Laboratory, Colorado State University Backstory Most summers, Coloradoans flock to the majestic Rocky Mountains with their beautiful hikes and various mountain activities. This is the case, at least, unless poor air quality forces them indoors.

But what about interactions; are any of those significant? I have heard some variant of this question from clinicians and researchers from many fields of science. While usually asked in earnest, this question is a dangerous one ; the sheer number of interactions can greatly inflate the number of false discoveries in the interactions, resulting in difficult-to-interpret models with many unnecessary interactions.

NoteReviewed by Logan Harris on 2025-11-12 Data Diction Data : things known or assumed as facts, making the basis of reasoning or calculation Diction : 1) the choice and use of words and phrases in speech or writing. 2) the choice of words especially with regard to correctness, clearness, or effectiveness.