From the course: Advanced Predictive Modeling: Mastering Ensembles and Metamodeling
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Error decomposition - SPSS Tutorial
From the course: Advanced Predictive Modeling: Mastering Ensembles and Metamodeling
Error decomposition
- [Instructor] Error decomposition is a fancy name, but this is going to be an important concept for us. Our models error has three components: variance, bias, and noise. The reason we care about this is that our motivation for using any ensemble technique is that it's either going to reduce variance or bias, or if we're lucky, it reduces both. Introduction to statistical learning is the classic text and it spends a whole chapter on this. I'm going to tell you everything you need to know for now, but if you really want to study this, you can't go wrong with this book. Especially if you're an R user. Don't worry if you're not an R user, it's the theory that counts. Now, if you do check it out, be prepared for an academic style book. Variance refers to the amount by which our model would change if we estimated it using a different training data set. Bias refers to the error that's introduced by approximating a real-life problem, which may be extremely complicated, by a much simpler…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.