Automatic Approximation of the Marginal Likelihood in Nonlinear Hierarchical Models

@inproceedings{Skaug2004AutomaticAO,
  title={Automatic Approximation of the Marginal Likelihood in Nonlinear Hierarchical Models},
  author={Hans J. Skaug and David A. Fournier},
  year={2004}
}
We show that the fitting of nonlinear hierarchical random effects models by maximum likelihood can be made automatic to the same extent that Bayesian model fitting can be automated by the program BUGS. The word ‘automatic’ here means that the technical details of computation are made transparent to the user. We achieve this by combining a technique from computer science known as ‘automatic differentiation’ with the Laplace approximation for calculating the marginal likelihood. Automatic… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 15 references

Maximum likelihood for generalized linear models with nested random effects via high-order, multivariate Laplace approximation

S. W. Raudenbush, M. L. Yang, M. Yosef
Journal of Computational and Graphical Statistics • 2000
View 4 Excerpts
Highly Influenced

The “SOCATT” dataset used in http://multilevel.ioe.ac.uk/softrev/tables.html

Ruppert
The classical “Polio” dataset (e.g. Kuk & Cheng • 2003

Automatic differentiation to facilitate maximum likelihood

H. J. Skaug
2002
View 2 Excerpts

Automatic differentiation to facilitate maximum likelihood estimation in nonlinear random effects models ’

H. J. Skaug
Journal of Computational and Graphical Statistics • 2002
View 2 Excerpts

An introduction to AD MODEL BUILDER Version 6.0.2 for use in nonlinear modeling and statistics, Available from http://otterrsch.com/admodel.htm

D. Fournier
2001
View 2 Excerpts

Mixed-Effects Models in S and S-PLUS, Statistics and Computing, Springer

J. C. Pinheiro, D. M. Bates
2000
View 1 Excerpt

Similar Papers

Loading similar papers…