Dataset Information

Automated Factor Slice Sampling.

ABSTRACT: Markov chain Monte Carlo (MCMC) algorithms offer a very general approach for sampling from arbitrary distributions. However, designing and tuning MCMC algorithms for each new distribution, can be challenging and time consuming. It is particularly difficult to create an efficient sampler when there is strong dependence among the variables in a multivariate distribution. We describe a two-pronged approach for constructing efficient, automated MCMC algorithms: (1) we propose the "factor slice sampler", a generalization of the univariate slice sampler where we treat the selection of a coordinate basis (factors) as an additional tuning parameter, and (2) we develop an approach for automatically selecting tuning parameters in order to construct an efficient factor slice sampler. In addition to automating the factor slice sampler, our tuning approach also applies to the standard univariate slice samplers. We demonstrate the efficiency and general applicability of our automated MCMC algorithm with a number of illustrative examples.

SUBMITTER: Tibbits MM

PROVIDER: S-EPMC4063214 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated Factor Slice Sampling.

Tibbits Matthew M MM Groendyke Chris C Haran Murali M Liechty John C JC

Journal of computational and graphical statistics : a joint publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America 20140101 2

Markov chain Monte Carlo (MCMC) algorithms offer a very general approach for sampling from arbitrary distributions. However, designing and tuning MCMC algorithms for each new distribution, can be challenging and time consuming. It is particularly difficult to create an efficient sampler when there is strong dependence among the variables in a multivariate distribution. We describe a two-pronged approach for constructing efficient, automated MCMC algorithms: (1) we propose the "factor slice sampl ...[more]

PMID: 24955002

Similar Datasets

Project description:Recent debates on insect decline require sound assessments on the relative drivers that may negatively impact insect populations. Often, baseline data rely on insect monitorings that integrate catches over long time periods. If, however, effects of time-critical environmental factors (e.g., light pollution) are of interest, higher temporal resolution of insect data is required during very specific time intervals (e.g., between dusk and dawn). Conventional time-critical insect trapping is labour-intensive (manual activation/deactivation) and temporally inaccurate as not all traps can be serviced synchronically at different sites. Also, temporal shifts of environmental conditions (e.g., sunset/sunrise) are not accounted for. We present a battery-driven automated insect flight-interception trap which samples insects during seven user-defined time intervals. A commercially available flight-interception trap is fitted to a turntable containing eight positions, seven of them holding cups and one consisting of a pass-through hole. While the cups sample insects during period of interest, the pass-through hole avoids unwanted sampling during time-intervals not of interest. Comparisons between two manual and two automated traps during 71 nights in 2018 showed no difference in caught insects. A study using 20 automated traps during 104 nights in 2019 proved that the automated flight-interception traps are reliable. The automated trap opens new research and application possibilities as arbitrary insect-sampling intervals can be defined. The trap proves efficient, saving manpower and associated costs as activation/deactivation is required only every seven sampling intervals. In addition, the timing of the traps is accurate, as all traps sample at exactly the same intervals and ensure comparability. The automated trap is low maintenance and robust due to straightforward technical design. It can be controlled manually or via smartphone through a Bluetooth connection. Full construction details are given in Appendices.

Dataset Information

Automated Factor Slice Sampling.

Publications

Automated Factor Slice Sampling.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets