EVRI-thing You Need to Know About Improving Models with Variable Selection

Professor Rasmus Bro of the University of Copenhagen presents.

Want more accurate predictions? Easier model interpretation? Sparse models? Less expensive measurements? There are many reasons for doing variable selection. In this webinar we will show how easy it is to do and how it can improve your models. Using different examples we will show how variable selection can make your models simpler and/or better.

Table of Contents

00:00 – Introduction and Outline

02:30 – Why use Variable Selection?

04:00 – Example: Beer Vis-NIR

14:20 – Why not use Variable Selection?

21:25 – Safeguard: Test Set

33:27 – Automatic Variable Selection

1:00:15 – Resources

Try it yourself. Get a free demo of Solo (or PLS_Toolbox) from https://www.software.eigenvector.com/toolbox/download/ (Free account creation required)