Linear and Non-linear Modeling

  • Andrew P. Robinson
  • Jeff D. Hamann
Part of the Use R book series (USE R)


This chapter describes some of the tools that are available in R for fitting certain kinds of conditional distributions; that is, constructing models to predict the behavior of one random variable given that the value of another one or more is known. Examples of such models in forestry include height-diameter models, diameter-volume models, and so on. Such models are of interest for two reasons: in order to make predictions and in order to estimate and interpret the parameters that describe the relationship. For example, a scientist might wish to know whether or not coring trees affects their growth and mortality, and how much; this problem is more naturally an interpretation and estimation problem. Alternatively, a manager might wish to predict heights for some trees for which only diameters and species are known; this problem is a prediction problem. The intended application of the model intimately affects the fitting process. Breiman (2001b) and the discussions that follow are excellent reading on this topic.


Nonlinear Regression Variance Function Interval Estimate Standardize Residual Search Path 
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 Science+Business Media, LLC 2011

Authors and Affiliations

  • Andrew P. Robinson
    • 1
  • Jeff D. Hamann
    • 2
  1. 1.Dept. Mathematics and StatisticsUniversity of MelbourneParkvilleAustralia
  2. 2.Forest Informatics, Inc.Corvallis OregonUSA

Personalised recommendations