You’re Naive if you think Linear Regression and Linear Regression are the same thing.

Yes, you read that right.

It starts with the difference between a statistical model and a machine learning model.

We understand that statistics are different from machine learning. But then why do we synonymously interchange a statistical linear regression model with that of a machine learning linear regression model? I assume its because they’re both coined ‘linear regression’. And that’s where we’re going wrong. At least not me anymore because I detest being wrong—ask my ex, and I’ll try my best to articulate the difference for you here.

Different Departments, Different Purposes

Linear regression can be used for two entirely different use-cases. That’s why we have both data analytics and predictive analytics in the first place. The first purpose of a linear regression model is entirely statistical—we try to understand the relationship between the independent variable and the dependent variable. This can look something like: “Will increasing ad spend by $1000 increase sales?". The second purpose of a linear regression model in the field of machine learning is to just make accurate predictions. That can look something like: “What will next month’s sales be?”