Generalized Regression Trees: Function Estimation via Recursive Partitioning and Maximum Likelihood

Abstract

A method that blends tree-structured nonparametric regression with classical maximum likelihood is used in a generalized regression setting. The function estimates constructed are piecewise polynomials and are produced together with decision trees containing useful information on the regressors. Fitting is carried out by applying maximum likelihood estimation to subsets of the data, where the subsets are selected via recursive partitioning and cross- validation pruning. Examples of Poisson and logistic regression trees are given to illustrate the method applied to count and binary response data. Large-sample properties of the estimates are derived under appropriate regularity conditions.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 25, 1993
Accession Number
ADA271259

Entities

People

  • Ching-ching Yang
  • Probal Chudhuri
  • Wei-yin Loh
  • Wen-da Lo

Organizations

  • University of Wisconsin–Madison

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Algorithms
  • Data Science
  • Diameters
  • Information Science
  • Maximum Likelihood Estimation
  • Observation
  • Polynomials
  • Probability
  • Random Variables
  • Residuals
  • Standards
  • Statistics
  • Terminals
  • Universities
  • Validation
  • Wisconsin

Fields of Study

  • Mathematics

Readers

  • Operations Research
  • Regression Analysis.