Cross Validation Although hypothesis tests provide a way to compare nested linear models, in many situations the approaches under consideration don't fit nicely in this paradigm. Indeed, for many modern tools and in many applications, the emphasis lies on prediction accuracy rather than on statistical significance. In these cases, cross validation provides a way to compare the predictive performance of competing methods. This is the second module in the Linear Models topic. Overview Learning Objectives Slide Deck Video Lecture Implement cross validation to assess the predictive value of a model using tools for iteration. Example I'll write code for today's content in a new R Markdown document called cross_validation.Rmd in the linear_models directory / repo. The code chunk below loads the usual packages (plus mgcv) and sets a seed for reproducibility. library(tidyverse) library(modelr) library(mgcv) library(p8105.datasets) set.seed(1)