Whether you were able to attend Part 1 of the webinar series "The Evolution of Regression Modeling: From Classical Linear Regression to Modern Ensembles" or not, the on-demand recording is now available. Watch the video and download the slides.
CART users often ask where they can find the value of the R‐squared for their regression trees. The answer is very simple: in conventional statistics.
Regression is one of the most popular modeling methods, but the classical approach has significant problems. This webinar series address these problems. Are you working with larger datasets? Is your data challenging? Does your data include missing values, nonlinear relationships, local patterns and interactions? This webinar series is for you! We will cover improvements to conventional and logistic regression, and will include a discussion of classical, regularized, and nonlinear regression, as well as modern ensemble and data mining approaches. This series will be of value to any classically trained statistician or modeler.
The parametric bootstrap methodology, introduced by Efron and Tibshirani in 1993 in their book "Introduction to the Bootstrap" is a remarkably simple and effective technique for exploring the variability in various model statistics and performance measures. When the subject matter is regression the method is especially easy to understand although it is often unintentionally obscured in technical statistical articles. Users of SPM 6 PRO EX or any of the 6 PRO EX series versions of CART and TreeNet already have access to the general bootstrap via the BATTERY mechanism (BATTERY BOOTSTRAP). In SPM 7 we also include the parametric bootstrap to help modelers better understand specific limits and inherent uncertainty in model characteristics and predictions. Here we focus only on regression; binary logistic models for the 0/1 classification problem will be discussed in another post.