A R Adaptive Multiple Importance Sampling (ARAMIS)
Luca Pozzi, Antonietta Mira · 2013
ARAMIS is an R package that runs the AMIS [1] algorithm. The main features of ARAMIS are parallelization and customization. ARAMIS exploits the massively parallel structure of AMIS to improve the performance of the algorithm as it was implemented in the original paper. As a result simulation time is reduced by orders of magnitudes. As for customization, the potential of the R language is fully exploited by ARAMIS which allows the user to taylor the software to the model which results from his or her own research setting. Target and proposal kernel can be easily specified by the user. Some working examples contained in the manual explain how this can be efficiently and easily done. As a consequence of the flexibility and efficiency of the package, even fairly complicated problems can be accommodated, e.g. sampling from an Extreme Value (EV) Copula distribution with a mixture of EV distributions as the proposal kernel. The latter is an interesting and useful example of how the user can specify some “real-world ” combination of target/proposal and it is added, in the manual, to the two working examples detailed in [1]. 1