From the course: Marketing Attribution and Mix Modeling

Unlock the full course today

Join today to access over 23,200 courses taught by industry experts.

Statistical tests to validate your model's accuracy

Statistical tests to validate your model's accuracy

From the course: Marketing Attribution and Mix Modeling

Start my 1-month free trial Buy for my team

Statistical tests to validate your model's accuracy

“

- [Narrator] How do you know you built an accurate model? Statisticians have developed a robust suite of test metrics you can calculate in order to tell if you're on the right path, we're looking at a marketing mix model that was built for an ice cream store, and we want to know how accurate it is. Let's start calculating the common statistical tests we can run. The first one is actually done for us by the LINEST function. So if we scroll down here, we can see the LINEST function right there. And this function is what's actually calculating our model coefficients, but as well as the model coefficients, it's also calculating the R squared. So we can see here at 0.97 and that's out of one. So this is a very high R squared. The next thing we want to know is are all of our variables statistically significant? And actually the LINEST function gives us everything we need there too. We just need to do further calculations.…

Contents