Classification and Regression trees can also be refereed to as CART for short. The origins of CART are somewhat difficult to pinpoint. Many people trace their locus back to Ronald Fisher‘s paper on discriminant analysis way back in 1936. And through the years traces of the algorithm were seen in various different publications. We even saw that in psychology, decision tree methods were used to model the human concept of learning. Somewhere along the way, researchers discovered that the algorithm was useful for programming and the CART algorithm started appearing in Automatic Interaction Detection work. AID work is essentially now one of the many components of Artificial Intelligence. From this work, University of Berkeley Statistics professors Leo Breiman and Charles Joel Stone, along with Jerome H. Friedman and Richard Olshen from Stanford University, began developing the classification and regression tree (CART) algorithm. This was unveiled in 1977. and in 1984 saw the first official publication with a CART software — a revolution in the world of algorithms.

Today, CART is one of the most used methods for data analysis. Along with upgrades to statistical software the algorithm has also expanded in its capacity for adaptability and accuracy, making it a very useful method of analysis. Though it is just one word, CART has an umbrella of many different types of analyses that can be used, but today we will focus on the one that we have used for the LNOB analysis