Model Selection and Multimodel Inference: A Practical Information-Theoretic ApproachWe wrote this book to introduce graduate students and research workers in various scienti?c disciplines to the use of information-theoretic approaches in the analysis of empirical data. These methods allow the data-based selection of a “best” model and a ranking and weighting of the remaining models in a pre-de?ned set. Traditional statistical inference can then be based on this selected best model. However, we now emphasize that information-theoretic approaches allow formal inference to be based on more than one model (m- timodel inference). Such procedures lead to more robust inferences in many cases, and we advocate these approaches throughout the book. The second edition was prepared with three goals in mind. First, we have tried to improve the presentation of the material. Boxes now highlight ess- tial expressions and points. Some reorganization has been done to improve the ?ow of concepts, and a new chapter has been added. Chapters 2 and 4 have been streamlined in view of the detailed theory provided in Chapter 7. S- ond, concepts related to making formal inferences from more than one model (multimodel inference) have been emphasized throughout the book, but p- ticularly in Chapters 4, 5, and 6. Third, new technical material has been added to Chapters 5 and 6. Well over 100 new references to the technical literature are given. These changes result primarily from our experiences while giving several seminars, workshops, and graduate courses on material in the ?rst e- tion. |
Contents
1 | 5 |
3 | 23 |
4 | 29 |
8 | 47 |
8 | 74 |
11 | 80 |
15 | 96 |
5 | 126 |
126 | 206 |
A Further Chain Binomial Model | 221 |
4 | 255 |
Advanced Issues and Deeper Insights | 267 |
4 | 289 |
Statistical Theory and Numerical Results | 352 |
8 | 429 |
Summary | 437 |
Other editions - View all
Common terms and phrases
AIC values AICc Akaike weights assumed Bayesian binomial bootstrap samples candidate models compute concept conditional confidence interval confidence set covariance criteria data analysis data dredging data set denoted derived effects evidence ratio example expected K-L fitted model function g₁ global model goodness-of-fit hence hypothesis testing important information-theoretic issue K-L best model K-L distance K-L information likelihood linear models linear regression log-likelihood log(g(x matrix methods model averaging model g model redundancy model selection uncertainty model set model structure models considered Monte Carlo notation number of parameters overdispersion overfitting parameter estimates parsimonious percentiles prediction predictor variables priori QAIC random variable relative sample size sampling variance Section selected model selection bias set of candidate set of models simple simulation statistical subset survival probabilities Table theoretical theory true model truth unconditional standard error var(ô variance-covariance matrix variation