BibTeX
@ARTICLE{
         Fournier2012AMB,
       author = "Fournier, David A. and Skaug, Hans J. and Ancheta, Johnoel and Ianelli, James and
         Magnusson, Arni and Maunder, Mark N. and Nielsen, Anders and Sibert, John",
       title = "{AD} {M}odel {B}uilder: using automatic differentiation for statistical inference of
         highly parameterized complex nonlinear models",
       journal = "Optimization Methods and Software",
       volume = "27",
       number = "2",
       pages = "233--249",
       year = "2012",
       doi = "10.1080/10556788.2011.597854",
       url = "http://www.tandfonline.com/doi/abs/10.1080/10556788.2011.597854",
       eprint = "http://www.tandfonline.com/doi/pdf/10.1080/10556788.2011.597854",
       abstract = "Many criteria for statistical parameter estimation, such as maximum likelihood, are
         formulated as a nonlinear optimization problem. Automatic Differentiation Model Builder (ADMB) is a
         programming framework based on automatic differentiation, aimed at highly nonlinear models with a
         large number of parameters. The benefits of using AD are computational efficiency and high numerical
         accuracy, both crucial in many practical problems. We describe the basic components and the
         underlying philosophy of ADMB, with an emphasis on functionality found in no other statistical
         software. One example of such a feature is the generic implementation of Laplace approximation of
         high-dimensional integrals for use in latent variable models. We also review the literature in which
         ADMB has been used, and discuss future development of ADMB as an open source project. Overall, the
         main advantages of ADMB are flexibility, speed, precision, stability and built-in methods to
         quantify uncertainty.",
       ad_tools = "AD Model Builder"
}