Richly parameterized linear models
additive, time series, and spatial models using random effects
Hodges, James S.
creator
text
bibliography
flu
Boca Raton
CRC Press
c2014
2014
monographic
eng
xxxviiii, 431 p. : illustrations ; 24 cm.
"This book covers a wide range of statistical models, including hierarchical, hierarchical generalized linear, linear mixed, dynamic linear, smoothing, spatial, and longitudinal. It presents a framework for expressing these richly parameterized models together as well as tools for exploring and interpreting the results of fitting the models to data. It extends the standard theory of linear models and illustrates the advantages and disadvantages of various theories. The book also examines surprising or undesirable results arising in the use of the models to analyze real data sets from collaborative research"--
I. Mixed Linear Models: Syntax, Theory, and Methods
1. An Opinionated Survey of Methods for Mixed Linear Models--
2. Two More Tools: Alternative Formulation, Measures of Complexity--
II Richly Parameterized Models as Mixed Linear Models--
3. Penalized Splines as Mixed Linear Models--
4. Additive Models and Models with Interactions--
5. Spatial Models as Mixed Linear Models--
6. Time-Series Models as Mixed Linear Models--
7. Two Other Syntaxes for Richly Parameterized Models--
III Form Linear Models to Richly Parameterized Models: Mean Structure--
8. Adapting Diagnostics from Linear Models--
9. Puzzles from Analyzing Real Datasets--
10. A Random Effect Competing with a Fixed Effect--
11. Differential Shrinkage--
12. Competition between Random Effects--
13. Random Effects Old and New--
Exercises--
James S. Hodges.
Includes bibliographical references (pages 413-424) and indexes.
Regression analysis
Linear models (Statistics)
MATHEMATICS / Probability & Statistics / General
000SA.062 H688
9781439866832 (hbk. : alk. paper)
ISI Library
130628
20141208123327.0
17795291