Exploring and Predicting using Linear Regression in R: Glossary

Make sure you set your working directory.

View a summary of your data to ensure your variables are of the right type.

There are different tests to test for correlation, depending on the type of variable.

Plots are useful to determine if there is a relationship, but this should be tested analytically.

Use helpful packages to visualise relationships.

Simple linear regression is for predicting the dependent variable Y based on one independent variable X.

R offers a large number of useful functions for performing linear regression.

Always check the model diagnostic plots and run the model diagnostic plots to check that the assumptions of linear regression are met.

Multiple linear regression is an intuitive extension of simple linear regression.

Transforming your data can give you a better fit.

Check for outliers, but ensure you take the correct course of action upon finding one.