#### DMCA

## A Note On the Unification of the Akaike Information Criterion (1998)

Venue: | J. Royal Statist. Soc. (B |

Citations: | 4 - 1 self |

### Citations

2900 |
Robust Statistics
- Huber
- 1981
(Show Context)
Citation Context ...amma fi )=2: (4) A reasonable criterion for judging the quality of candidate models with respect to the data is E 0 f\Delta( fi;soe)g, where fi andsoe are the M-estimators of fi and oe, respectively (=-=Huber, 1981-=-, p. 176). Since, in practice, E 0 f\Delta(sfi;soe)g cannot be computed, we obtain two versions of its approximation, which are given in the following theorem. Theorem. Under above three assumptions, ... |

616 | Robust estimation of a location parameter - Huber - 1964 |

512 |
Statistical Models and Methods for Lifetime Data
- Lawless
- 1982
(Show Context)
Citation Context ... AICC Let aef(y i \Gamma x 0 i fi)=oeg = \Gamma2 ln[oe \Gamma1 gf(y i \Gamma x 0 i fi)=oeg], where g(t) = exp(\Gammat) exp (\Gammae \Gammat ) and is the density of the extreme value distribution (see =-=Lawless, 1982-=-, p. 298). After simple calculations we find that R / 2 (t)g(t)d t ( R /(t)g(t)d t) 2 = 1; and Zs/( toe 0soe )g(t)d t = 2 Z exp(\Gamma toe 0soe )g(t)d t = 2\Gamma(1 + oe 0soe ): In addition, \Delta(fi... |

441 |
Regression and time series model selection in small samples
- Hurvich, Tsai
- 1989
(Show Context)
Citation Context ...ch unify many versions of the Akaike information criterion family. These unified criteria can be extended to various model structures with appropriate robust functions, such as autoregressive models (=-=Hurvich and Tsai, 1989-=-), multivariate regression models (Bedrick and Tsai, 1994), extended quasi-likelihood models (Hurvich and Tsai, 1995), and others. In addition, the expectations in (5) and (6) remain applicable to the... |

160 | Some comments on - Mallows - 1973 |

38 |
Model selection for multivariate regression in small samples
- Bedrick, Tsai
- 1994
(Show Context)
Citation Context ...n family. These unified criteria can be extended to various model structures with appropriate robust functions, such as autoregressive models (Hurvich and Tsai, 1989), multivariate regression models (=-=Bedrick and Tsai, 1994-=-), extended quasi-likelihood models (Hurvich and Tsai, 1995), and others. In addition, the expectations in (5) and (6) remain applicable to the location-scale family as long as the explanatory variabl... |

27 |
Model Selection for Extended Quasi-Likelihood Models in Small Samples. [research-article]. Biometrics(3), 1077. doi: 10.2307/2533006 IBM SPSS Statistics 21. (2012). IBM SPSS Statistics 21 Core Systems User's Guide: IBM Corporations. Retrieved from ftp://p
- Hurvich, Tsai
- 1995
(Show Context)
Citation Context ... model structures with appropriate robust functions, such as autoregressive models (Hurvich and Tsai, 1989), multivariate regression models (Bedrick and Tsai, 1994), extended quasi-likelihood models (=-=Hurvich and Tsai, 1995-=-), and others. In addition, the expectations in (5) and (6) remain applicable to the location-scale family as long as the explanatory variables x 1 ; :::; x n (Section 2) are identically distributed a... |

17 | Improved estimators of KullbackLeibler information for autoregressive model selection in small samples - Hurvich, Shumway, et al. - 1990 |

17 | Robust Model Selection in Regression - Ronchetti - 1985 |

12 | Model selection for least absolute deviations regression in small samples. Statistics & probability Letters - Hurvich, Tsai - 1990 |

10 | Asymptotic behavior of general M-estimates for regression and scale with random carriers - Maronna, Yohai - 1981 |

5 | Asymptotic behavior of robust estimates of regression and scale parameters with fixed carriers - Silvapulle - 1985 |

4 | A robust version of Mallows - Ronchetti, Staudte - 1994 |