Model Robust Confidence Intervals Using Maximum Likelihood Estimators
Richard M. Royall · International Statistical Review · 1986
La «methode delta» fournit un estimateur alternatif simple de la variance qui demeure consistant sous des conditions plus generales et qui donne par consequent des intervalles de confiance robustes
Cited by 578 works. doi:10.2307/1403146
Authors
References
- The behavior of maximum likelihood estimates under nonstandard conditions
- Quasi-Likelihood Functions, Generalized Linear Models, and the Gauss-Newton Method
- Quasi-Likelihood Functions
- Tests of Separate Families of Hypotheses
- Assessing the Accuracy of the Maximum Likelihood Estimator: Observed Versus Expected Fisher Information
- Some Problems Connected with Statistical Inference
- Some Remarks on Overdispersion
- On Power Transformations to Symmetry
- Jackknifing in Unbalanced Situations
- Robust properties of likelihood ratio tests