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.

How to deal with high-order multiple nominals

How to deal with high-order multiple nominals

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

Start my 1-month free trial

How to deal with high-order multiple nominals

“

- [Instructor] Now let's talk about an issue that I think very few modelers handle in an optimal way. Let's take a look at the far right-hand side of the phone service customers data set, specifically the phone model variable, and I've turned the filter on so that we can quickly take a look at how many models we have. What most analysts would do at this point is just treat this as a nominal variable, but it simply is not the best way to go. For one, there's quite a few categories here. It's not at the breaking point, but there's quite a few. So you could debate whether or not to call this a high order nominal or a very high order nominal, but what are we going to accomplish if we put this directly into the model? What's going to happen when these model numbers change? There's a much better way, and let's take a look at what it is. Now I have a new data set, the phone model spreadsheet, and this is found in the originals…

Contents