Adaptive Regression by Mixing

Thumbnail Image
Date
1999
Authors
Yang, Yuhong
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Authors
Research Projects
Organizational Units
Organizational Unit
Statistics
As leaders in statistical research, collaboration, and education, the Department of Statistics at Iowa State University offers students an education like no other. We are committed to our mission of developing and applying statistical methods, and proud of our award-winning students and faculty.
Journal Issue
Is Version Of
Versions
Series
Department
Abstract

Adaptation over different procedures is of practical importance. Different procedures perform well under different conditions. In many practical situations, it is rather hard to assess which conditions are (approximately) satisfied so as to identify the best procedure for the data at hand. Thus automatic adaptation over various scenarios is desirable. A practically feasible method, named adaptive regression by mixing (ARM), is proposed to convexly combine general candidate regression procedures. Under mild conditions, the resulting estimator is theoretically shown to perform optimally in rates of convergence without knowing which of the original procedures work the best. Simulations are conducted in several settings, including comparing a parametric model with nonparametric alternatives, comparing a neural network with a projection pursuit in multidimensional regression, and combining bandwidths in kernel regression. The results clearly support the theoretical property of ARM. The ARM algorithm assigns weights on the candidate models-procedures via proper assessment of performance of the estimators. The data are split into two parts, one for estimation and the other for measuring behavior in prediction. Although there are many plausible ways to assign the weights, ARM has a connection with information theory, which ensures the desired adaptation capability. Indeed, under mild conditions, we show that the squared L2 risk of the estimator based on ARM is basically bounded above by the risk of each candidate procedure plus a small penalty term of order 1/n. Minimizing over the procedures gives the automatically optimal rate of convergence for ARM. Model selection often induces unnecessarily large variability in estimation. Alternatively, a proper weighting of the candidate models can be more stable, resulting in a smaller risk. Simulations suggest that ARM works better than model selection using Akaike or Bayesian information criteria when the error variance is not very small.

Comments

This preprint was published as Yuhong Yang, "Adaptive Regression by Mixing", Journal of the American Statistical Association (2001): 574-588, doi: 10.1198/016214501753168262.

Description
Keywords
Citation
DOI
Source
Subject Categories
Copyright
Collections