After a thorough revision that removed most of the theoretical attempts at improving our understanding of AMIS convergence, we have now resubmitted the AMIS paper to Scandinavian Journal of Statistics and arXived the new version as well. (I remind the reader that AMIS stands for adaptive mixture importance sampling and that it implements an adaptive version of Owen and Zhou’s (2000, JASA) stabilisation mixture technique, using this correction on the past and present importance weights, at each iteration of this iterative algorithm.) The AMIS method starts being used in population genetics, including an on-going work by Jean-Marie Cornuet and a published paper in Molecular Biology and Evolution by Sirén, Marttinen and Corander. The challenge of properly demonstrating AMIS convergence remains open!
Filed under: R, Statistics, University life Tagged: adaptive importance sampling, adaptive mixture importance sampling, AMIS, IMIS