AIExplainer
Machine Learning Intermediate 1 min read

What is R-squared?

A statistical measure that evaluates the goodness of fit of a model

R-squared measures how well a model explains the variation in a dataset, with higher values indicating a better fit

Think of R-squared like a report card for your model, where 100% would mean the model perfectly predicts the data, and 0% means it doesn't explain anything

A company uses R-squared to evaluate a model that predicts sales based on advertising spend, finding that the model explains 80% of the variation in sales

R-squared is used to evaluate the performance of regression models, helping to identify which factors are most important in predicting an outcome

R-squared does not measure the accuracy of predictions, but rather how well the model fits the data, and high R-squared values do not necessarily mean the model is useful or generalizable

The concept of R-squared was first introduced by statistician Karl Pearson in the early 20th century

coefficient of determination goodness of fit

Three products for different needs — explore what’s relevant to you.