Taylor, Paul C. and Jones, M. C.
|DOI (Digital Object Identifier) Link:||http://dx.doi.org/10.1080/00949659608811771|
|Google Scholar:||Look up in Google Scholar|
Breiman, Friedman, Olshen and Stone (1984) expounded a method called Classification and Regression Trees, or CART, which can be used for nonparametric discrimination and regression. Taylor and Silverman (1993) presented a new splitting criterion for the growing of classification trees; this new criterion was called the mean posterior improvement criterion. This paper extends the mean posterior improvement criterion to the case of regression trees. The extension is made via kernel density estimation. General results on how to select the bandwidth (smoothing parameter) appropriate to estimation of the mean posterior improvement criterion are obtained. These results are adapted to allow a practical implementation of the mean posterior improvement criterion for regression trees. Examples of the behaviour of the new criterion relative to currently used splitting criteria are given.
|Item Type:||Journal Article|
|Copyright Holders:||1996 Taylor and Francis|
|Keywords:||CART; mean posterior improvement criterion; kernel density estimation; estimation of functional of mixtures of distributions; bandwidth selection|
|Academic Unit/Department:||Mathematics, Computing and Technology > Mathematics and Statistics
Mathematics, Computing and Technology
|Depositing User:||Sarah Frain|
|Date Deposited:||29 Mar 2011 12:02|
|Last Modified:||18 Jan 2016 10:05|
|Share this page:|