9. Statistical Modelling

  1. Michael J. Crawley

Published Online: 6 NOV 2012

DOI: 10.1002/9781118448908.ch9

Book Title

The R Book, Second Edition

The R Book, Second Edition

Additional Information

How to Cite

Crawley, M. J. (2012) Statistical Modelling, in The R Book, Second Edition, John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9781118448908.ch9

Author Information

  1. Imperial College London at Silwood Park, UK

Publication History

  1. Published Online: 6 NOV 2012
  2. Published Print: 7 DEC 2012

ISBN Information

Print ISBN: 9780470973929

Online ISBN: 9781118448908

SEARCH

CHAPTER TOOLS

Get PDF (389K)

Keywords:

  • statistical modelling;
  • principle of parsimony (Occam's razor);
  • model simplification steps, caveats;
  • model formulae in R, creating formula objects;
  • update function in model simplification;
  • model checking, heteroscedasticity/non-normality of errors;
  • summary of statistical models in R;
  • optional arguments in model-fitting functions;
  • Akaike's information criterion;
  • model simplification by stepwise deletion, treatment /Helmert/Sum

Summary

This chapter contains sections titled:

  • First things first

  • Maximum likelihood

  • The principle of parsimony (Occam's razor)

  • Types of statistical model

  • Steps involved in model simplification

  • Model formulae in R

  • Multiple error terms

  • The intercept as parameter 1

  • The update function in model simplification

  • Model formulae for regression

  • Box-Cox transformations

  • Model criticism

  • Model checking

  • Influence

  • Summary of statistical models in R

  • Optional arguments in model-fitting functions

  • Akaike's information criterion

  • Leverage

  • Misspecified model

  • Model checking in R

  • Extracting information from model objects

  • The summary tables for continuous and categorical explanatory variables

  • Contrasts

  • Model simplification by stepwise deletion

  • Comparison of the three kinds of contrasts

  • Aliasing

  • Orthogonal polynomial contrasts: contr.poly

  • Summary of statistical modeling

Get PDF (389K)