From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Anscombe's quartet

Anscombe's quartet

From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Start my 1-month free trial

Anscombe's quartet

“

- [Instructor] Okay, let's talk about Anscombe's quartet. This data, which you can find in the Originals folder, was developed by a statistician named Francis Anscombe way back in 1973. It has become justifiably famous as a cautionary tale. What he's trying to caution us about is don't pay attention only to descriptive statistics, and the absence of a graphical representation. So here's the trick. To understand what this demonstration is all about, you want to treat these various pairs almost as if they're separate datasets. We have an X1, Y1 pair, X2, Y2, X3, Y3 and so on. So let's first do some basic descriptive statistics. We can keep it very basic indeed. I'll go ahead and calculate an average here and then I'll close off the parens and we see that we get an average. And I'm just going to drag that across so that we can get an average for everything. And what do we have? We've got an average of nine on all of…

Contents