New Wave Nonparametrics

  • Bertrand Clarke
  • Ernest Fokoué
  • Hao Helen Zhang
Part of the Springer Series in Statistics book series (SSS)

By the late 1980 s, Classical nonparametrics was established as “classical”. Concurrently, however, the beginnings of a different stream of nonparametric thinking were already under way. Indeed, its origins go back to the 1970 s if not earlier. The focus here is not on large spaces of functions but on classes of functions intended to be tractable representations for intermediate tranches. The models retain much of the flexibility of Classical methods but are much more interpretable; not as interpretable as many subject matter specialists might want but possessing much more structure than themethods of Chapters 2 and 3. In practice, computer-intensive procedures pervade these more recent techniques. This permits iterative fitting algorithms, cross-validation for model selection, bootstrapping for pointwise confidence bands on the estimated functions as seen earlier, and much more besides.


Hide Layer Additive Model Regression Function Terminal Node Generalize Additive Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag New York 2009

Authors and Affiliations

  • Bertrand Clarke
    • 1
  • Ernest Fokoué
    • 2
  • Hao Helen Zhang
    • 3
  1. 1.University of MiamiMiamiCanada
  2. 2.Department of Science & MathematicsKettering UniversityFlintUSA
  3. 3.Department of StatisticsNorth Carolina State University Program in Statistical GeneticsRaleighUSA

Personalised recommendations