Maximum Likelihood Regression Trees
Xiaogang SuMorgan WangJuanjuan Fan
Abstract
We propose a method of constructing regression trees within the framework of maximum likelihood. It inherits the backward fitting idea of classification and regression trees (CART) but has more rigorous justification. Simulation studies show that it provides more accurate tree model selection compared to CART. The analysis of a baseball dataset is given as an illustration.
Metrics
Citation History
Topics
Related Documents
Maximum Likelihood Ridge Regression
Maximum Likelihood and Nonlinear Regression
Maximum Likelihood and Nonlinear Regression
Asymmetric Laplace Regression: Maximum Likelihood, Maximum Entropy and Quantile Regression
Anil K. BeraAntonio F. GalvaoGabriel Montes‐RojasSung Y. Park
Maximum likelihood trees for “internal segments”.
Michelle Wille (122037)Victoria Grillo (558452)Silvia Ban de Gouvea Pedroso (12525063)Graham W. Burgess (12525066)Allison Crawley (12525069)Celia Dickason (12525072)Philip M. Hansbro (7501037)Md. Ahasanul Hoque (9774491)Paul F. Horwood (8103170)Peter D. Kirkland (11459896)Nina Yu-Hsin Kung (12525075)Stacey E. Lynch (9204496)Sue Martin (12525078)Michaela McArthur (12525081)Kim O’Riley (3587468)Andrew J. Read (7487945)Simone Warner (2825933)Bethany J. Hoye (12525084)Simeon Lisovski (3271050)Trent Leen (12525087)Aeron C. Hurt (8103167)Jeff Butler (546551)Ivano Broz (145849)Kelly R. Davies (12525090)Patrick Mileto (12525093)Matthew J. Neave (11554970)Vicky Stevens (12525096)Andrew C. Breed (11258169)Tommy T. Y. Lam (12525099)Edward C. Holmes (8067413)Marcel Klaassen (23343)Frank Y. K. Wong (12525102)