Support > White Papers

White Papers

1. An Overview of CART® Methodology


• Here is a brief introduction and overview of the methodology.
• Text only


2. Technical Note for Statisticians


• Some technical aspects of CART® analyses are of special interest to statisticians; we list the most important ones here.
• Text only


3. Do Splitting Rules Really Matter?


• Contrary to popular opinion in data mining circles, our experience indicates that splitting criteria do matter; in fact, that the difference between using the right rule and the wrong rule could add up to millions of dollars of lost opportunity.
• Text only


4. MARS® (Multivariate Adaptive Regression Splines)


• WHAT IS "MARS"? MARS, is a multivariate non-parametric regression procedure introduced in 1991 by world-renowned Stanford statistician and physicist, Jerome Friedman (Friedman, 1991). Learn more here.
• Text only


5. Hybrid CART-Logit Model in Classification & Data Mining


• We introduce a new method for combining CART and logit, which exhibits performance advantages, and admits of a natural set of statistical evaluation tests.
• (PowerPoint presentation available for download) - 120k


6. Achieving Results With Next Generation Data Mining Techniques


• How do you get all you can from your data mining projects?
• (PowerPoint presentation available for download) - 218k


7. Forecasting Recessions: Can We Do Better on MARS?


• The purpose of this paper is to revisit the information contained in the financial variables using non-linear, nonparametric methods, in particular MARS (Multivariate Adaptive Regression Splines) for the purpose of forecasting regression.
• PDF file available for download.


8. Shares, Bonds, or Cash? Asset allocation in the new economy, using CART


• How an Australian stock brokerage is using CART to optimize allocation of investments.
• PDF file available for download - 158k


9. Critical Features of High Performance Decision Trees


• This document reviews some of key features of decision trees that any informed analyst should be thinking about when choosing this kind of data mining tool for important data analyses.
• PDF file available for download - 158k


10. Statistical Process Analysis of Medical Incidents


• by Norio Suzuki, Sojiro Kirihara, and Atsushi Ootaki of Meiji University, Japan
• PDF file available for download - 127k


11. Data Mining Approaches to Modeling Insurance Risk


• by Inna Kolyshkina and Richard Brookes of PricewaterhouseCoopers
• This paper discusses the use of some data mining techniques in insurance and presents two case studies illustrating the application of classification and regression trees (CART), multivariate adaptive regression splines (MARS) and hybrid models.
• PDF file available for download - 260k


12. CART, MARS, TreeNet® , and Neural Networks


• An overview of how CART, MARS, and TreeNet® can be of value follows.
• Text only


13. Artificial Intelligence and Data Mining


• Short discussion on Artificial Intelligence and Data Mining (Machine Learning).
• Text only


14. Data Warehousing and Data Mining


• If you have managed to construct a data warehouse, you now have corporate and operational data organized to support informed decision making. The challenge is to find effective ways to analyze those data so as to extract valuable information.
• Text only


15. Customer Relationship Management and Data Mining


• Every successful marketing enterprise devotes considerable effort to customer relationship management (CRM).
• Text only


16. Martian Chronicles: Is MARS better than Neural Networks?


• by Louise Francis, FCAS, MAAA
• This paper introduces MARS by showing its similarity to linear regression, and then applying it to insurance fraud data. It then compares its performance to that of neural networks.
• PDF file available for download - 500k