https://github.com/helske/stannis

Importance sampling correction approach for faster inference for exponential family state space models
https://github.com/helske/stannis

Last synced: 6 days ago
JSON representation

Importance sampling correction approach for faster inference for exponential family state space models

• Host: GitHub
• URL: https://github.com/helske/stannis
• Owner: helske
• Created: 2017-04-01T09:08:37.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-08-08T15:44:11.000Z (over 7 years ago)
• Last Synced: 2024-10-29T09:18:53.757Z (about 2 months ago)
• Language: C++
• Size: 745 KB
• Stars: 4
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# stannis: R package combining Stan and the efficient importance sampling type correction MCMC for Poisson time series

This R package provides a small comparison between different MCMC algorithms for the Bayesian inference of structural time series with Poisson distributed observations, and implements a hybrid algorithm using Stan with combination with importance sampling type correction of MCMC suggested by Vihola, Helske, and Franks (2016) (https://arxiv.org/abs/1609.02541).

See vignette for details, and [the poster from UseR!2017](https://github.com/helske/stannis/blob/master/stannis.pdf) for quick preview.